Datasets:

NeurIPSConference

/

NeurIPS2024-papers

like 0

Follow

Conference on Neural Information Processing Systems 1

https://openreview.net/forum?id=taI8M5DiXj

@inproceedings{ ghoummaid2024when, title={When to Act and When to Ask: Policy Learning With Deferral Under Hidden Confounding}, author={Marah Ghoummaid and Uri Shalit}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=taI8M5DiXj} }

We consider the task of learning how to act in collaboration with a human expert based on observational data. The task is motivated by high-stake scenarios such as healthcare and welfare where algorithmic action recommendations are made to a human expert, opening the option of deferring making a recommendation in cases where the human might act better on their own. This task is especially challenging when dealing with observational data, as using such data runs the risk of hidden confounders whose existence can lead to biased and harmful policies. However, unlike standard policy learning, the presence of a human expert can mitigate some of these risks. We build on the work of Mozannar and Sontag (2020) on consistent surrogate loss for learning with the option of deferral to an expert, where they solve a cost-sensitive supervised classification problem. Since we are solving a causal problem, where labels don’t exist, we use a causal model to learn costs which are robust to a bounded degree of hidden confounding. We prove that our approach can take advantage of the strengths of both the model and the expert to obtain a better policy than either. We demonstrate our results by conducting experiments on synthetic and semi-synthetic data and show the advantages of our method compared to baselines.

When to Act and When to Ask: Policy Learning With Deferral Under Hidden Confounding

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tZtepJBtHg

@inproceedings{ h{\"u}botter2024transductive, title={Transductive Active Learning: Theory and Applications}, author={Jonas H{\"u}botter and Bhavya Sukhija and Lenart Treven and Yarden As and Andreas Krause}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tZtepJBtHg} }

We study a generalization of classical active learning to real-world settings with concrete prediction targets where sampling is restricted to an accessible region of the domain, while prediction targets may lie outside this region. We analyze a family of decision rules that sample adaptively to minimize uncertainty about prediction targets. We are the first to show, under general regularity assumptions, that such decision rules converge uniformly to the smallest possible uncertainty obtainable from the accessible data. We demonstrate their strong sample efficiency in two key applications: active fine-tuning of large neural networks and safe Bayesian optimization, where they achieve state-of-the-art performance.

Transductive Active Learning: Theory and Applications

NeurIPS.cc/2024/Conference

2402.15898

poster

https://openreview.net/forum?id=tZRpvLXevU

@inproceedings{ boutin2024latent, title={Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks}, author={Victor Boutin and Rishav Mukherji and Aditya Agrawal and Sabine Muzellec and Thomas FEL and Thomas Serre and Rufin VanRullen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tZRpvLXevU} }

Humans can effortlessly draw new categories from a single exemplar, a feat that has long posed a challenge for generative models. However, this gap has started to close with recent advances in diffusion models. This one-shot drawing task requires powerful inductive biases that have not been systematically investigated. Here, we study how different inductive biases shape the latent space of Latent Diffusion Models (LDMs). Along with standard LDM regularizers (KL and vector quantization), we explore supervised regularizations (including classification and prototype-based representation) and contrastive inductive biases (using SimCLR and redundancy reduction objectives). We demonstrate that LDMs with redundancy reduction and prototype-based regularizations produce near-human-like drawings (regarding both samples' recognizability and originality) -- better mimicking human perception (as evaluated psychophysically). Overall, our results suggest that the gap between humans and machines in one-shot drawings is almost closed.

Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks

NeurIPS.cc/2024/Conference

2406.06079

poster

https://openreview.net/forum?id=tYdR1lTWqh

@inproceedings{ ji2024reversing, title={Reversing the Forget-Retain Objectives: An Efficient {LLM} Unlearning Framework from Logit Difference}, author={Jiabao Ji and Yujian Liu and Yang Zhang and Gaowen Liu and Ramana Rao Kompella and Sijia Liu and Shiyu Chang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tYdR1lTWqh} }

As Large Language Models (LLMs) demonstrate extensive capability in learning from documents, LLM unlearning becomes an increasingly important research area to address concerns of LLMs in terms of privacy, copyright, etc. A conventional LLM unlearning task typically involves two goals: (1) The target LLM should forget the knowledge in the specified forget documents; and (2) it should retain the other knowledge that the LLM possesses, for which we assume access to a small number of retain documents. To achieve both goals, a mainstream class of LLM unlearning methods introduces an optimization framework with a combination of two objectives – maximizing the prediction loss on the forget documents while minimizing that on the retain documents, which suffers from two challenges, degenerated output and catastrophic forgetting. In this paper, we propose a novel unlearning framework called Unlearning from Logit Difference (ULD), which introduces an assistant LLM that aims to achieve the opposite of the unlearning goals: remembering the forget documents and forgetting the retain knowledge. ULD then derives the unlearned LLM by computing the logit difference between the target and the assistant LLMs. We show that such reversed objectives would naturally resolve both aforementioned challenges while significantly improving the training efficiency. Extensive experiments demonstrate that our method efficiently achieves the intended forgetting while preserving the LLM’s overall capabilities, reducing training time by more than threefold. Notably, our method loses 0% of model utility on the ToFU benchmark, whereas baseline methods may sacrifice 17% of utility on average to achieve comparable forget quality.

Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference

NeurIPS.cc/2024/Conference

2406.08607

https://huggingface.co/papers/2406.08607

poster

https://openreview.net/forum?id=tWkL7k1u5v

@inproceedings{ pertigkiozoglou2024improving, title={Improving Equivariant Model Training via Constraint Relaxation}, author={Stefanos Pertigkiozoglou and Evangelos Chatzipantazis and Shubhendu Trivedi and Kostas Daniilidis}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tWkL7k1u5v} }

Equivariant neural networks have been widely used in a variety of applications due to their ability to generalize well in tasks where the underlying data symmetries are known. Despite their successes, such networks can be difficult to optimize and require careful hyperparameter tuning to train successfully. In this work, we propose a novel framework for improving the optimization of such models by relaxing the hard equivariance constraint during training: We relax the equivariance constraint of the network's intermediate layers by introducing an additional non-equivariant term that we progressively constrain until we arrive at an equivariant solution. By controlling the magnitude of the activation of the additional relaxation term, we allow the model to optimize over a larger hypothesis space containing approximate equivariant networks and converge back to an equivariant solution at the end of training. We provide experimental results on different state-of-the-art network architectures, demonstrating how this training framework can result in equivariant models with improved generalization performance. Our code is available at https://github.com/StefanosPert/Equivariant_Optimization_CR

Improving Equivariant Model Training via Constraint Relaxation

NeurIPS.cc/2024/Conference

2408.13242

poster

https://openreview.net/forum?id=tVConYid20

@inproceedings{ shah2024flashattention, title={FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision}, author={Jay Shah and Ganesh Bikshandi and Ying Zhang and Vijay Thakkar and Pradeep Ramani and Tri Dao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tVConYid20} }

Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. elaborated an approach to speed up attention on GPUs through minimizing memory reads/writes. However, it has yet to take advantage of new capabilities present in recent hardware, with FlashAttention-2 achieving only 35% utilization on the H100 GPU. We develop three main techniques to speed up attention on Hopper GPUs: exploiting asynchrony of the Tensor Cores and TMA to (1) overlap overall computation and data movement via warp-specialization and (2) interleave block-wise matmul and softmax operations, and (3) block quantization and incoherent processing that leverages hardware support for FP8 low-precision. We demonstrate that our method, FlashAttention-3, achieves speedup on H100 GPUs by 1.5-2.0$\times$ with BF16 reaching up to 840 TFLOPs/s (85\% utilization), and with FP8 reaching 1.3 PFLOPs/s. We validate that FP8 FlashAttention-3 achieves 2.6$\times$ lower numerical error than a baseline FP8 attention.

FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision

NeurIPS.cc/2024/Conference

2407.08608

https://huggingface.co/papers/2407.08608

oral

https://openreview.net/forum?id=tUpcRQNvVM

@inproceedings{ bhatt2024deep, title={Deep Submodular Peripteral Networks}, author={Gantavya Bhatt and Arnav Mohanty Das and Jeff Bilmes}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tUpcRQNvVM} }

Submodular functions, crucial for various applications, often lack practical learning methods for their acquisition. Seemingly unrelated, learning a scaling from oracles offering graded pairwise preferences (GPC) is underexplored, despite a rich history in psychometrics. In this paper, we introduce deep submodular peripteral networks (DSPNs), a novel parametric family of submodular functions, and methods for their training using a GPC-based strategy to connect and then tackle both of the above challenges. We introduce newly devised GPC-style ``peripteral'' loss which leverages numerically graded relationships between pairs of objects (sets in our case). Unlike traditional contrastive learning, or RHLF preference ranking, our method utilizes graded comparisons, extracting more nuanced information than just binary-outcome comparisons, and contrasts sets of any size (not just two). We also define a novel suite of automatic sampling strategies for training, including active-learning inspired submodular feedback. We demonstrate DSPNs' efficacy in learning submodularity from a costly target submodular function and demonstrate its superiority both for experimental design and online streaming applications.

Deep Submodular Peripteral Networks

NeurIPS.cc/2024/Conference

2403.08199

oral

https://openreview.net/forum?id=tUHABDZP0Q

@inproceedings{ xu2024reinforced, title={Reinforced Cross-Domain Knowledge Distillation on Time Series Data}, author={QING XU and Min Wu and Xiaoli Li and Kezhi Mao and Zhenghua Chen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tUHABDZP0Q} }

Unsupervised domain adaptation methods have demonstrated superior capabilities in handling the domain shift issue which widely exists in various time series tasks. However, their prominent adaptation performances heavily rely on complex model architectures, posing an unprecedented challenge in deploying them on resource-limited devices for real-time monitoring. Existing approaches, which integrates knowledge distillation into domain adaptation frameworks to simultaneously address domain shift and model complexity, often neglect network capacity gap between teacher and student and just coarsely align their outputs over all source and target samples, resulting in poor distillation efficiency. Thus, in this paper, we propose an innovative framework named Reinforced Cross-Domain Knowledge Distillation (RCD-KD) which can effectively adapt to student's network capability via dynamically selecting suitable target domain samples for knowledge transferring. Particularly, a reinforcement learning-based module with a novel reward function is proposed to learn optimal target sample selection policy based on student's capacity. Meanwhile, a domain discriminator is designed to transfer the domain invariant knowledge. Empirical experimental results and analyses on four public time series datasets demonstrate the effectiveness of our proposed method over other state-of-the-art benchmarks.

Reinforced Cross-Domain Knowledge Distillation on Time Series Data

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tTpVHsqTKf

@inproceedings{ zheng2024syncvis, title={Sync{VIS}: Synchronized Video Instance Segmentation}, author={rongkun Zheng and Lu Qi and Xi Chen and Yi Wang and Kun Wang and Yu Qiao and Hengshuang Zhao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tTpVHsqTKf} }

Recent DETR-based methods have advanced the development of Video Instance Segmentation (VIS) through transformers' efficiency and capability in modeling spatial and temporal information. Despite harvesting remarkable progress, existing works follow asynchronous designs, which model video sequences via either video-level queries only or adopting query-sensitive cascade structures, resulting in difficulties when handling complex and challenging video scenarios. In this work, we analyze the cause of this phenomenon and the limitations of the current solutions, and propose to conduct synchronized modeling via a new framework named SyncVIS. Specifically, SyncVIS explicitly introduces video-level query embeddings and designs two key modules to synchronize video-level query with frame-level query embeddings: a synchronized video-frame modeling paradigm and a synchronized embedding optimization strategy. The former attempts to promote the mutual learning of frame- and video-level embeddings with each other and the latter divides large video sequences into small clips for easier optimization. Extensive experimental evaluations are conducted on the challenging YouTube-VIS 2019 & 2021 & 2022, and OVIS benchmarks, and SyncVIS achieves state-of-the-art results, which demonstrates the effectiveness and generality of the proposed approach. The code is available at https://github.com/rkzheng99/SyncVIS.

SyncVIS: Synchronized Video Instance Segmentation

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tTnFH7D1h4

@inproceedings{ heng2024outofdistribution, title={Out-of-Distribution Detection with a Single Unconditional Diffusion Model}, author={Alvin Heng and Alexandre H. Thiery and Harold Soh}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tTnFH7D1h4} }

Out-of-distribution (OOD) detection is a critical task in machine learning that seeks to identify abnormal samples. Traditionally, unsupervised methods utilize a deep generative model for OOD detection. However, such approaches require a new model to be trained for each inlier dataset. This paper explores whether a single model can perform OOD detection across diverse tasks. To that end, we introduce Diffusion Paths (DiffPath), which uses a single diffusion model originally trained to perform unconditional generation for OOD detection. We introduce a novel technique of measuring the rate-of-change and curvature of the diffusion paths connecting samples to the standard normal. Extensive experiments show that with a single model, DiffPath is competitive with prior work using individual models on a variety of OOD tasks involving different distributions. Our code is publicly available at https://github.com/clear-nus/diffpath.

Out-of-Distribution Detection with a Single Unconditional Diffusion Model

NeurIPS.cc/2024/Conference

2405.11881

poster

https://openreview.net/forum?id=tSWoT8ttkO

@inproceedings{ luo2024efficient, title={Efficient Recurrent Off-Policy {RL} Requires a Context-Encoder-Specific Learning Rate}, author={Fan-Ming Luo and Zuolin Tu and Zefang Huang and Yang Yu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tSWoT8ttkO} }

Real-world decision-making tasks are usually partially observable Markov decision processes (POMDPs), where the state is not fully observable. Recent progress has demonstrated that recurrent reinforcement learning (RL), which consists of a context encoder based on recurrent neural networks (RNNs) for unobservable state prediction and a multilayer perceptron (MLP) policy for decision making, can mitigate partial observability and serve as a robust baseline for POMDP tasks. However, prior recurrent RL algorithms have faced issues with training instability. In this paper, we find that this instability stems from the autoregressive nature of RNNs, which causes even small changes in RNN parameters to produce large output variations over long trajectories. Therefore, we propose **R**ecurrent Off-policy RL with Context-**E**ncoder-**S**p**e**cific **L**earning Rate (RESeL) to tackle this issue. Specifically, RESeL uses a lower learning rate for context encoder than other MLP layers to ensure the stability of the former while maintaining the training efficiency of the latter. We integrate this technique into existing off-policy RL methods, resulting in the RESeL algorithm. We evaluated RESeL in 18 POMDP tasks, including classic, meta-RL, and credit assignment scenarios, as well as five MDP locomotion tasks. The experiments demonstrate significant improvements in training stability with RESeL. Comparative results show that RESeL achieves notable performance improvements over previous recurrent RL baselines in POMDP tasks, and is competitive with or even surpasses state-of-the-art methods in MDP tasks. Further ablation studies highlight the necessity of applying a distinct learning rate for the context encoder. Code is available at https://github.com/FanmingL/Recurrent-Offpolicy-RL.

Efficient Recurrent Off-Policy RL Requires a Context-Encoder-Specific Learning Rate

NeurIPS.cc/2024/Conference

2405.15384

poster

https://openreview.net/forum?id=tRRWoa9e80

@inproceedings{ hu2024token, title={Token Merging for Training-Free Semantic Binding in Text-to-Image Synthesis}, author={taihang Hu and Linxuan Li and Joost van de Weijer and Hongcheng Gao and Fahad Khan and Jian Yang and Ming-Ming Cheng and Kai Wang and Yaxing Wang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tRRWoa9e80} }

Although text-to-image (T2I) models exhibit remarkable generation capabilities, they frequently fail to accurately bind semantically related objects or attributes in the input prompts; a challenge termed semantic binding. Previous approaches either involve intensive fine-tuning of the entire T2I model or require users or large language models to specify generation layouts, adding complexity. In this paper, we define semantic binding as the task of associating a given object with its attribute, termed attribute binding, or linking it to other related sub-objects, referred to as object binding. We introduce a novel method called Token Merging (ToMe), which enhances semantic binding by aggregating relevant tokens into a single composite token. This ensures that the object, its attributes and sub-objects all share the same cross-attention map. Additionally, to address potential confusion among main objects with complex textual prompts, we propose end token substitution as a complementary strategy. To further refine our approach in the initial stages of T2I generation, where layouts are determined, we incorporate two auxiliary losses, an entropy loss and a semantic binding loss, to iteratively update the composite token to improve the generation integrity. We conducted extensive experiments to validate the effectiveness of ToMe, comparing it against various existing methods on the T2I-CompBench and our proposed GPT-4o object binding benchmark. Our method is particularly effective in complex scenarios that involve multiple objects and attributes, which previous methods often fail to address. The code will be publicly available at https://github.com/hutaihang/ToMe

Token Merging for Training-Free Semantic Binding in Text-to-Image Synthesis

NeurIPS.cc/2024/Conference

2411.07132

poster

https://openreview.net/forum?id=tQukGCDaNT

@inproceedings{ yin2024improved, title={Improved Distribution Matching Distillation for Fast Image Synthesis}, author={Tianwei Yin and Micha{\"e}l Gharbi and Taesung Park and Richard Zhang and Eli Shechtman and Fredo Durand and William T. Freeman}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tQukGCDaNT} }

Recent approaches have shown promises distilling expensive diffusion models into efficient one-step generators. Amongst them, Distribution Matching Distillation (DMD) produces one-step generators that match their teacher in distribution, i.e., the distillation process does not enforce a one-to-one correspondence with the sampling trajectories of their teachers. However, to ensure stable training in practice, DMD requires an additional regression loss computed using a large set of noise--image pairs, generated by the teacher with many steps of a deterministic sampler. This is not only computationally expensive for large-scale text-to-image synthesis, but it also limits the student's quality, tying it too closely to the teacher's original sampling paths. We introduce DMD2, a set of techniques that lift this limitation and improve DMD training. First, we eliminate the regression loss and the need for expensive dataset construction. We show that the resulting instability is due to the "fake" critic not estimating the distribution of generated samples with sufficient accuracy and propose a two time-scale update rule as a remedy. Second, we integrate a GAN loss into the distillation procedure, discriminating between generated samples and real images. This lets us train the student model on real data, thus mitigating the imperfect "real" score estimation from the teacher model, and thereby enhancing quality. Third, we introduce a new training procedure that enables multi-step sampling in the student, and addresses the training--inference input mismatch of previous work, by simulating inference-time generator samples during training. Taken together, our improvements set new benchmarks in one-step image generation, with FID scores of 1.28 on ImageNet-64×64 and 8.35 on zero-shot COCO 2014, surpassing the original teacher despite a 500X reduction in inference cost. Further, we show our approach can generate megapixel images by distilling SDXL, demonstrating exceptional visual quality among few-step methods, and surpassing the teacher. We release our code and pretrained models.

Improved Distribution Matching Distillation for Fast Image Synthesis

NeurIPS.cc/2024/Conference

2405.14867

https://huggingface.co/papers/2405.14867

oral

https://openreview.net/forum?id=tPgagXpvcV

@inproceedings{ krzakala2024anygraph, title={Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss}, author={Paul KRZAKALA and Junjie Yang and R{\'e}mi Flamary and Florence d'Alch{\'e}-Buc and Charlotte Laclau and Matthieu Labeau}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tPgagXpvcV} }

We propose Any2graph, a generic framework for end-to-end Supervised Graph Prediction (SGP) i.e. a deep learning model that predicts an entire graph for any kind of input. The framework is built on a novel Optimal Transport loss, the Partially-Masked Fused Gromov-Wasserstein, that exhibits all necessary properties (permutation invariance, differentiability and scalability) and is designed to handle any-sized graphs. Numerical experiments showcase the versatility of the approach that outperform existing competitors on a novel challenging synthetic dataset and a variety of real-world tasks such as map construction from satellite image (Sat2Graph) or molecule prediction from fingerprint (Fingerprint2Graph).

Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss

NeurIPS.cc/2024/Conference

2402.12269

oral

https://openreview.net/forum?id=tPdJ2qHkOB

@inproceedings{ tian2024toward, title={Toward Self-Improvement of {LLM}s via Imagination, Searching, and Criticizing}, author={Ye Tian and Baolin Peng and Linfeng Song and Lifeng Jin and Dian Yu and Lei Han and Haitao Mi and Dong Yu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tPdJ2qHkOB} }

Despite the impressive capabilities of Large Language Models (LLMs) on various tasks, they still struggle with scenarios that involves complex reasoning and planning. Self-correction and self-learning emerge as viable solutions, employing strategies that allow LLMs to refine their outputs and learn from self-assessed rewards. Yet, the efficacy of LLMs in self-refining its response, particularly in complex reasoning and planning task, remains dubious. In this paper, we introduce AlphaLLM for the self-improvements of LLMs, which integrates Monte Carlo Tree Search (MCTS) with LLMs to establish a self-improving loop, thereby enhancing the capabilities of LLMs without additional annotations. Drawing inspiration from the success of AlphaGo, AlphaLLM addresses the unique challenges of combining MCTS with LLM for self-improvement, including data scarcity, the vastness search spaces of language tasks, and the subjective nature of feedback in language tasks. AlphaLLM is comprised of prompt synthesis component, an efficient MCTS approach tailored for language tasks, and a trio of critic models for precise feedback. Our experimental results in mathematical reasoning tasks demonstrate that AlphaLLM significantly enhances the performance of LLMs without additional annotations, showing the potential for self-improvement in LLMs.

Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing

NeurIPS.cc/2024/Conference

2404.12253

https://huggingface.co/papers/2404.12253

poster

https://openreview.net/forum?id=tOXoQPRzPL

@inproceedings{ yu2024an, title={An Image is Worth 32 Tokens for Reconstruction and Generation}, author={Qihang Yu and Mark Weber and Xueqing Deng and Xiaohui Shen and Daniel Cremers and Liang-Chieh Chen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tOXoQPRzPL} }

Recent advancements in generative models have highlighted the crucial role of image tokenization in the efficient synthesis of high-resolution images. Tokenization, which transforms images into latent representations, reduces computational demands compared to directly processing pixels and enhances the effectiveness and efficiency of the generation process. Prior methods, such as VQGAN, typically utilize 2D latent grids with fixed downsampling factors. However, these 2D tokenizations face challenges in managing the inherent redundancies present in images, where adjacent regions frequently display similarities. To overcome this issue, we introduce **T**ransformer-based 1-D**i**mensional **Tok**enizer (TiTok), an innovative approach that tokenizes images into 1D latent sequences. TiTok provides a more compact latent representation, yielding substantially more efficient and effective representations than conventional techniques. For example, a 256 × 256 × 3 image can be reduced to just **32** discrete tokens, a significant reduction from the 256 or 1024 tokens obtained by prior methods. Despite its compact nature, TiTok achieves competitive performance to state-of-the-art approaches. Specifically, using the same generator framework, TiTok attains **1.97** gFID, outperforming MaskGIT baseline significantly by 4.21 at ImageNet 256 × 256 benchmark. The advantages of TiTok become even more significant when it comes to higher resolution. At ImageNet 512 × 512 benchmark, TiTok not only outperforms state-of-the-art diffusion model DiT-XL/2 (gFID 2.74 vs. 3.04), but also reduces the image tokens by 64×, leading to **410× faster** generation process. Our best-performing variant can significantly surpasses DiT-XL/2 (gFID **2.13** vs. 3.04) while still generating high-quality samples **74× faster**. Codes and models are available at https://github.com/bytedance/1d-tokenizer

An Image is Worth 32 Tokens for Reconstruction and Generation

NeurIPS.cc/2024/Conference

2406.07550

https://huggingface.co/papers/2406.07550

poster

https://openreview.net/forum?id=tNhwg9U767

@inproceedings{ wang2024microstructures, title={Microstructures and Accuracy of Graph Recall by Large Language Models}, author={Yanbang Wang and Hejie Cui and Jon Kleinberg}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tNhwg9U767} }

Graphs data is crucial for many applications, and much of it exists in the relations described in textual format. As a result, being able to accurately recall and encode a graph described in earlier text is a basic yet pivotal ability that LLMs need to demonstrate if they are to perform reasoning tasks that involve graph-structured information. Human performance at graph recall by has been studied by cognitive scientists for decades, and has been found to often exhibit certain structural patterns of bias that align with human handling of social relationships. To date, however, we know little about how LLMs behave in analogous graph recall tasks: do their recalled graphs also exhibit certain biased patterns, and if so, how do they compare with humans and affect other graph reasoning tasks? In this work, we perform the first systematical study of graph recall by LLMs, investigating the accuracy and biased microstructures (local structural patterns) in their recall. We find that LLMs not only underperform often in graph recall, but also tend to favor more triangles and alternating 2-paths. Moreover, we find that more advanced LLMs have a striking dependence on the domain that a real-world graph comes from --- by yielding the best recall accuracy when the graph is narrated in a language style consistent with its original domain.

Microstructures and Accuracy of Graph Recall by Large Language Models

NeurIPS.cc/2024/Conference

2402.11821

poster

https://openreview.net/forum?id=tN0xnYPLt6

@inproceedings{ li2024tinylut, title={Tiny{LUT}: Tiny Look-Up Table for Efficient Image Restoration at the Edge}, author={Huanan LI and Juntao Guan and Lai Rui and Sijun Ma and Lin Gu and Zhangming Zhu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tN0xnYPLt6} }

Look-up tables(LUTs)-based methods have recently shown enormous potential in image restoration tasks, which are capable of significantly accelerating the inference. However, the size of LUT exhibits exponential growth with the convolution kernel size, creating a storage bottleneck for its broader application on edge devices. Here, we address the storage explosion challenge to promote the capacity of mapping the complex CNN models by LUT. We introduce an innovative separable mapping strategy to achieve over $7\times$ storage reduction, transforming the storage from exponential dependence on kernel size to a linear relationship. Moreover, we design a dynamic discretization mechanism to decompose the activation and compress the quantization scale that further shrinks the LUT storage by $4.48\times$. As a result, the storage requirement of our proposed TinyLUT is around 4.1\% of MuLUT-SDY-X2 and amenable to on-chip cache, yielding competitive accuracy with over $5\times$ lower inference latency on Raspberry 4B than FSRCNN. Our proposed TinyLUT enables superior inference speed on edge devices with new state-of-the-art accuracy on both of image super-resolution and denoising, showcasing the potential of applying this method to various image restoration tasks at the edge. The codes are available at: https://github.com/Jonas-KD/TinyLUT.

TinyLUT: Tiny Look-Up Table for Efficient Image Restoration at the Edge

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tLXgzQ5WZl

@inproceedings{ ren2024scube, title={{SC}ube: Instant Large-Scale Scene Reconstruction using VoxSplats}, author={Xuanchi Ren and Yifan Lu and hanxue liang and Jay Zhangjie Wu and Huan Ling and Mike Chen and Sanja Fidler and Francis Williams and Jiahui Huang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tLXgzQ5WZl} }

We present SCube, a novel method for reconstructing large-scale 3D scenes (geometry, appearance, and semantics) from a sparse set of posed images. Our method encodes reconstructed scenes using a novel representation VoxSplat, which is a set of 3D Gaussians supported on a high-resolution sparse-voxel scaffold. To reconstruct a VoxSplat from images, we employ a hierarchical voxel latent diffusion model conditioned on the input images followed by a feedforward appearance prediction model. The diffusion model generates high-resolution grids progressively in a coarse-to-fine manner, and the appearance network predicts a set of Gaussians within each voxel. From as few as 3 non-overlapping input images, SCube can generate millions of Gaussians with a 10243 voxel grid spanning hundreds of meters in 20 seconds. Past works tackling scene reconstruction from images either rely on per-scene optimization and fail to reconstruct the scene away from input views (thus requiring dense view coverage as input) or leverage geometric priors based on low-resolution models, which produce blurry results. In contrast, SCube leverages high-resolution sparse networks and produces sharp outputs from few views. We show the superiority of SCube compared to prior art using the Waymo self-driving dataset on 3D reconstruction and demonstrate its applications, such as LiDAR simulation and text-to-scene generation.

SCube: Instant Large-Scale Scene Reconstruction using VoxSplats

NeurIPS.cc/2024/Conference

2410.20030

poster

https://openreview.net/forum?id=tLWoxftJVh

@inproceedings{ li2024consistency, title={Consistency Purification: Effective and Efficient Diffusion Purification towards Certified Robustness}, author={Yiquan Li and Zhongzhu Chen and Kun Jin and Jiongxiao Wang and Jiachen Lei and Bo Li and Chaowei Xiao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tLWoxftJVh} }

Diffusion Purification, purifying noised images with diffusion models, has been widely used for enhancing certified robustness via randomized smoothing. However, existing frameworks often grapple with the balance between efficiency and effectiveness. While the Denoising Diffusion Probabilistic Model (DDPM) offers an efficient single-step purification, it falls short in ensuring purified images reside on the data manifold. Conversely, the Stochastic Diffusion Model effectively places purified images on the data manifold but demands solving cumbersome stochastic differential equations, while its derivative, the Probability Flow Ordinary Differential Equation (PF-ODE), though solving simpler ordinary differential equations, still requires multiple computational steps. In this work, we demonstrated that an ideal purification pipeline should generate the purified images on the data manifold that are as much semantically aligned to the original images for effectiveness in one step for efficiency. Therefore, we introduced Consistency Purification, an efficiency-effectiveness Pareto superior purifier compared to the previous work. Consistency Purification employs the consistency model, a one-step generative model distilled from PF-ODE, thus can generate on-manifold purified images with a single network evaluation. However, the consistency model is designed not for purification thus it does not inherently ensure semantic alignment between purified and original images. To resolve this issue, we further refine it through Consistency Fine-tuning with LPIPS loss, which enables more aligned semantic meaning while keeping the purified images on data manifold. Our comprehensive experiments demonstrate that our Consistency Purification framework achieves state-of-the-art certified robustness and efficiency compared to baseline methods.

Consistency Purification: Effective and Efficient Diffusion Purification towards Certified Robustness

NeurIPS.cc/2024/Conference

2407.00623

poster

https://openreview.net/forum?id=tKuLgnDWWN

@inproceedings{ cai2024silence, title={{SILENCE}: Protecting privacy in offloaded speech understanding on resource-constrained devices}, author={DONGQI CAI and Shangguang Wang and Zeling Zhang and Felix Xiaozhu Lin and Mengwei Xu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tKuLgnDWWN} }

Speech serves as a ubiquitous input interface for embedded mobile devices. Cloud-based solutions, while offering powerful speech understanding services, raise significant concerns regarding user privacy. To address this, disentanglement-based encoders have been proposed to remove sensitive information from speech signals without compromising the speech understanding functionality. However, these encoders demand high memory usage and computation complexity, making them impractical for resource-constrained wimpy devices. Our solution is based on a key observation that speech understanding hinges on long-term dependency knowledge of the entire utterance, in contrast to privacy-sensitive elements that are short-term dependent. Exploiting this observation, we propose SILENCE, a lightweight system that selectively obscuring short-term details, without damaging the long-term dependent speech understanding performance. The crucial part of SILENCE is a differential mask generator derived from interpretable learning to automatically configure the masking process. We have implemented SILENCE on the STM32H7 microcontroller and evaluate its efficacy under different attacking scenarios. Our results demonstrate that SILENCE offers speech understanding performance and privacy protection capacity comparable to existing encoders, while achieving up to 53.3$\times$ speedup and 134.1$\times$ reduction in memory footprint.

SILENCE: Protecting privacy in offloaded speech understanding on resource-constrained devices

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tJGX7tpGO8

@inproceedings{ li2024what, title={What Matters in Graph Class Incremental Learning? An Information Preservation Perspective}, author={Jialu Li and Yu Wang and Pengfei Zhu and Wanyu Lin and Qinghua Hu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tJGX7tpGO8} }

Graph class incremental learning (GCIL) requires the model to classify emerging nodes of new classes while remembering old classes. Existing methods are designed to preserve effective information of old models or graph data to alleviate forgetting, but there is no clear theoretical understanding of what matters in information preservation. In this paper, we consider that present practice suffers from high semantic and structural shifts assessed by two devised shift metrics. We provide insights into information preservation in GCIL and find that maintaining graph information can preserve information of old models in theory to calibrate node semantic and graph structure shifts. We correspond graph information into low-frequency local-global information and high-frequency information in spatial domain. Based on the analysis, we propose a framework, Graph Spatial Information Preservation (GSIP). Specifically, for low-frequency information preservation, the old node representations obtained by inputting replayed nodes into the old model are aligned with the outputs of the node and its neighbors in the new model, and then old and new outputs are globally matched after pooling. For high-frequency information preservation, the new node representations are encouraged to imitate the near-neighbor pair similarity of old node representations. GSIP achieves a 10\% increase in terms of the forgetting metric compared to prior methods on large-scale datasets. Our framework can also seamlessly integrate existing replay designs. The code is available through https://github.com/Jillian555/GSIP.

What Matters in Graph Class Incremental Learning? An Information Preservation Perspective

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tIzW3l2uaN

@inproceedings{ li2024onetonormal, title={One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection}, author={Yiyue Li and Shaoting Zhang and Kang Li and Qicheng Lao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tIzW3l2uaN} }

Traditional Anomaly Detection (AD) methods have predominantly relied on unsupervised learning from extensive normal data. Recent AD methods have evolved with the advent of large pre-trained vision-language models, enhancing few-shot anomaly detection capabilities. However, these latest AD methods still exhibit limitations in accuracy improvement. One contributing factor is their direct comparison of a query image's features with those of few-shot normal images. This direct comparison often leads to a loss of precision and complicates the extension of these techniques to more complex domains—an area that remains underexplored in a more refined and comprehensive manner. To address these limitations, we introduce the anomaly personalization method, which performs a personalized one-to-normal transformation of query images using an anomaly-free customized generation model, ensuring close alignment with the normal manifold. Moreover, to further enhance the stability and robustness of prediction results, we propose a triplet contrastive anomaly inference strategy, which incorporates a comprehensive comparison between the query and generated anomaly-free data pool and prompt information. Extensive evaluations across eleven datasets in three domains demonstrate our model's effectiveness compared to the latest AD methods. Additionally, our method has been proven to transfer flexibly to other AD methods, with the generated image data effectively improving the performance of other AD methods.

One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tGozvLTDY3

@inproceedings{ xu2024dgslam, title={{DG}-{SLAM}: Robust Dynamic Gaussian Splatting {SLAM} with Hybrid Pose Optimization}, author={Yueming Xu and Haochen Jiang and Zhongyang Xiao and Jianfeng Feng and Li Zhang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tGozvLTDY3} }

Achieving robust and precise pose estimation in dynamic scenes is a significant research challenge in Visual Simultaneous Localization and Mapping (SLAM). Recent advancements integrating Gaussian Splatting into SLAM systems have proven effective in creating high-quality renderings using explicit 3D Gaussian models, significantly improving environmental reconstruction fidelity. However, these approaches depend on a static environment assumption and face challenges in dynamic environments due to inconsistent observations of geometry and photometry. To address this problem, we propose DG-SLAM, the first robust dynamic visual SLAM system grounded in 3D Gaussians, which provides precise camera pose estimation alongside high-fidelity reconstructions. Specifically, we propose effective strategies, including motion mask generation, adaptive Gaussian point management, and a hybrid camera tracking algorithm to improve the accuracy and robustness of pose estimation. Extensive experiments demonstrate that DG-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, and novel-view synthesis in dynamic scenes, outperforming existing methods meanwhile preserving real-time rendering ability.

DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization

NeurIPS.cc/2024/Conference

2411.08373

poster

https://openreview.net/forum?id=tGDUDKirAy

@inproceedings{ wu2024verified, title={Verified Safe Reinforcement Learning for Neural Network Dynamic Models}, author={Junlin Wu and Huan Zhang and Yevgeniy Vorobeychik}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tGDUDKirAy} }

Learning reliably safe autonomous control is one of the core problems in trustworthy autonomy. However, training a controller that can be formally verified to be safe remains a major challenge. We introduce a novel approach for learning verified safe control policies in nonlinear neural dynamical systems while maximizing overall performance. Our approach aims to achieve safety in the sense of finite-horizon reachability proofs, and is comprised of three key parts. The first is a novel curriculum learning scheme that iteratively increases the verified safe horizon. The second leverages the iterative nature of gradient-based learning to leverage incremental verification, reusing information from prior verification runs. Finally, we learn multiple verified initial-state-dependent controllers, an idea that is especially valuable for more complex domains where learning a single universal verified safe controller is extremely challenging. Our experiments on five safe control problems demonstrate that our trained controllers can achieve verified safety over horizons that are as much as an order of magnitude longer than state-of-the-art baselines, while maintaining high reward, as well as a perfect safety record over entire episodes.

Verified Safe Reinforcement Learning for Neural Network Dynamic Models

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tFB5SsabVb

@inproceedings{ mercatali2024graph, title={Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series}, author={Giangiacomo Mercatali and Andre Freitas and Jie Chen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tFB5SsabVb} }

Interacting systems are prevalent in nature. It is challenging to accurately predict the dynamics of the system if its constituent components are analyzed independently. We develop a graph-based model that unveils the systemic interactions of time series observed at irregular time points, by using a directed acyclic graph to model the conditional dependencies (a form of causal notation) of the system components and learning this graph in tandem with a continuous-time model that parameterizes the solution curves of ordinary differential equations (ODEs). Our technique, a graph neural flow, leads to substantial enhancements over non-graph-based methods, as well as graph-based methods without the modeling of conditional dependencies. We validate our approach on several tasks, including time series classification and forecasting, to demonstrate its efficacy.

Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series

NeurIPS.cc/2024/Conference

2410.14030

poster

https://openreview.net/forum?id=tEEpVPDaRf

@inproceedings{ jang2024identity, title={Identity Decoupling for Multi-Subject Personalization of Text-to-Image Models}, author={Sangwon Jang and Jaehyeong Jo and Kimin Lee and Sung Ju Hwang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tEEpVPDaRf} }

Text-to-image diffusion models have shown remarkable success in generating personalized subjects based on a few reference images. However, current methods often fail when generating multiple subjects simultaneously, resulting in mixed identities with combined attributes from different subjects. In this work, we present MuDI, a novel framework that enables multi-subject personalization by effectively decoupling identities from multiple subjects. Our main idea is to utilize segmented subjects generated by a foundation model for segmentation (Segment Anything) for both training and inference, as a form of data augmentation for training and initialization for the generation process. Moreover, we further introduce a new metric to better evaluate the performance of our method on multi-subject personalization. Experimental results show that our MuDI can produce high-quality personalized images without identity mixing, even for highly similar subjects as shown in Figure 1. Specifically, in human evaluation, MuDI obtains twice the success rate for personalizing multiple subjects without identity mixing over existing baselines and is preferred over 70% against the strongest baseline.

Identity Decoupling for Multi-Subject Personalization of Text-to-Image Models

NeurIPS.cc/2024/Conference

2404.04243

https://huggingface.co/papers/2404.04243

poster

https://openreview.net/forum?id=tDvFa5OJyS

@inproceedings{ wenger2024computationaware, title={Computation-Aware Gaussian Processes: Model Selection And Linear-Time Inference}, author={Jonathan Wenger and Kaiwen Wu and Philipp Hennig and Jacob R. Gardner and Geoff Pleiss and John Patrick Cunningham}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tDvFa5OJyS} }

Model selection in Gaussian processes scales prohibitively with the size of the training dataset, both in time and memory. While many approximations exist, all incur inevitable approximation error. Recent work accounts for this error in the form of computational uncertainty, which enables---at the cost of quadratic complexity---an explicit tradeoff between computation and precision. Here we extend this development to model selection, which requires significant enhancements to the existing approach, including linear-time scaling in the size of the dataset. We propose a novel training loss for hyperparameter optimization and demonstrate empirically that the resulting method can outperform SGPR, CGGP and SVGP, state-of-the-art methods for GP model selection, on medium to large-scale datasets. Our experiments show that model selection for computation-aware GPs trained on 1.8 million data points can be done within a few hours on a single GPU. As a result of this work, Gaussian processes can be trained on large-scale datasets without significantly compromising their ability to quantify uncertainty---a fundamental prerequisite for optimal decision-making.

Computation-Aware Gaussian Processes: Model Selection And Linear-Time Inference

NeurIPS.cc/2024/Conference

2411.01036

poster

https://openreview.net/forum?id=tDMTwto6jv

@inproceedings{ gao2024selbald, title={{SEL}-{BALD}: Deep Bayesian Active Learning for Selective Labeling with Instance Rejection}, author={Ruijiang Gao and Mingzhang Yin and Maytal Saar-Tsechansky}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tDMTwto6jv} }

Machine learning systems are widely used in many high-stakes contexts in which experimental designs for assigning treatments are infeasible. When evaluating a decision instance is costly, such as investigating a fraud case, or evaluating a biopsy decision, a sample-efficient strategy is needed. However, while existing active learning methods assume humans will always label the instances selected by the machine learning model, in many critical applications, humans may decline to label instances selected by the machine learning model due to reasons such as regulation constraint, domain knowledge, or algorithmic aversion, thus not sample efficient. In this paper, we propose the Active Learning with Instance Rejection (ALIR) problem, which is a new active learning problem that considers the human discretion behavior for high-stakes decision making problems. We propose new active learning algorithms under deep Bayesian active learning for selective labeling (SEL-BALD) to address the ALIR problem. Our algorithms consider how to acquire information for both the machine learning model and the human discretion model. We conduct experiments on both synthetic and real-world datasets to demonstrate the effectiveness of our proposed algorithms.

SEL-BALD: Deep Bayesian Active Learning for Selective Labeling with Instance Rejection

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tCf7S75xFa

@inproceedings{ hamelijnck2024physicsinformed, title={Physics-Informed Variational State-Space Gaussian Processes}, author={Oliver Hamelijnck and Arno Solin and Theodoros Damoulas}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tCf7S75xFa} }

Differential equations are important mechanistic models that are integral to many scientific and engineering applications. With the abundance of available data there has been a growing interest in data-driven physics-informed models. Gaussian processes (GPs) are particularly suited to this task as they can model complex, non-linear phenomena whilst incorporating prior knowledge and quantifying uncertainty. Current approaches have found some success but are limited as they either achieve poor computational scalings or focus only on the temporal setting. This work addresses these issues by introducing a variational spatio-temporal state-space GP that handles linear and non-linear physical constraints while achieving efficient linear-in-time computation costs. We demonstrate our methods in a range of synthetic and real-world settings and outperform the current state-of-the-art in both predictive and computational performance.

Physics-Informed Variational State-Space Gaussian Processes

NeurIPS.cc/2024/Conference

2409.13876

poster

https://openreview.net/forum?id=tBRNC6YemY

@inproceedings{ patil2024gorilla, title={Gorilla: Large Language Model Connected with Massive {API}s}, author={Shishir G Patil and Tianjun Zhang and Xin Wang and Joseph E. Gonzalez}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tBRNC6YemY} }

Large Language Models (LLMs) have seen an impressive wave of advances, with models now excelling in a variety of tasks, such as mathematical reasoning and program synthesis. However, their potential to effectively use tools via API calls remains unfulfilled. This is a challenging task even for today’s state-of-the-art LLMs such as GPT-4 largely due to their unawareness of what APIs are available and how to use them in a frequently updated tool set. We develop Gorilla, a finetuned LLaMA model that surpasses the performance of GPT-4 on writing API calls. Trained with the novel Retriever Aware Training (RAT), when combined with a document retriever, Gorilla demonstrates a strong capability to adapt to test-time document changes, allowing flexible user updates or version changes. It also substantially mitigates the issue of hallucination, commonly encountered when prompting LLMs directly. To evaluate the model’s ability, we introduce APIBench, a comprehensive dataset consisting of HuggingFace, TorchHub, and TensorHub APIs. The successful integration of the retrieval system with Gorilla demonstrates the potential for LLMs to use tools more accurately, keep up with frequently updated documentation, and consequently increase the reliability and applicability of their outputs. Gorilla’s code, model, data, and demo are available at: https://gorilla.cs.berkeley.edu

Gorilla: Large Language Model Connected with Massive APIs

NeurIPS.cc/2024/Conference

2305.15334

https://huggingface.co/papers/2305.15334

poster

https://openreview.net/forum?id=tAlMAcqK9s

@inproceedings{ aliakbarpour2024optimal, title={Optimal Algorithms for Augmented Testing of Discrete Distributions}, author={Maryam Aliakbarpour and Piotr Indyk and Ronitt Rubinfeld and Sandeep Silwal}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tAlMAcqK9s} }

We consider the problem of hypothesis testing for discrete distributions. In the standard model, where we have sample access to an underlying distribution $p$, extensive research has established optimal bounds for uniformity testing, identity testing (goodness of fit), and closeness testing (equivalence or two-sample testing). We explore these problems in a setting where a predicted data distribution, possibly derived from historical data or predictive machine learning models, is available. We demonstrate that such a predictor can indeed reduce the number of samples required for all three property testing tasks. The reduction in sample complexity depends directly on the predictor’s quality, measured by its total variation distance from $p$. A key advantage of our algorithms is their adaptability to the precision of the prediction. Specifically, our algorithms can self-adjust their sample complexity based on the accuracy of the available prediction, operating without any prior knowledge of the estimation’s accuracy (i.e. they are consistent). Additionally, we never use more samples than the standard approaches require, even if the predictions provide no meaningful information (i.e. they are also robust). We provide lower bounds to indicate that the improvements in sample complexity achieved by our algorithms are information-theoretically optimal. Furthermore, experimental results show that the performance of our algorithms on real data significantly exceeds our worst-case guarantees for sample complexity, demonstrating the practicality of our approach.

Optimal Algorithms for Augmented Testing of Discrete Distributions

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=tAOg1HdvGy

@inproceedings{ chen2024interpolating, title={Interpolating Item and User Fairness in Multi-Sided Recommendations}, author={Qinyi Chen and Jason Cheuk Nam Liang and Negin Golrezaei and Djallel Bouneffouf}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=tAOg1HdvGy} }

Today's online platforms heavily lean on algorithmic recommendations for bolstering user engagement and driving revenue. However, these recommendations can impact multiple stakeholders simultaneously---the platform, items (sellers), and users (customers)---each with their unique objectives, making it difficult to find the right middle ground that accommodates all stakeholders. To address this, we introduce a novel fair recommendation framework, Problem (FAIR), that flexibly balances multi-stakeholder interests via a constrained optimization formulation. We next explore Problem (FAIR) in a dynamic online setting where data uncertainty further adds complexity, and propose a low-regret algorithm FORM that concurrently performs real-time learning and fair recommendations, two tasks that are often at odds. Via both theoretical analysis and a numerical case study on real-world data, we demonstrate the efficacy of our framework and method in maintaining platform revenue while ensuring desired levels of fairness for both items and users.

Interpolating Item and User Fairness in Multi-Sided Recommendations

NeurIPS.cc/2024/Conference

2306.10050

poster

https://openreview.net/forum?id=t9gNEhreht

@inproceedings{ li2024selma, title={{SELMA}: Learning and Merging Skill-Specific Text-to-Image Experts with Auto-Generated Data}, author={Jialu Li and Jaemin Cho and Yi-Lin Sung and Jaehong Yoon and Mohit Bansal}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t9gNEhreht} }

Recent text-to-image (T2I) generation models have demonstrated impressive capabilities in creating images from text descriptions. However, these T2I generation models often fail to generate images that precisely match the details of the text inputs, such as incorrect spatial relationship or missing objects. In this paper, we introduce SELMA: Skill-Specific Expert Learning and Merging with Auto-Generated Data, a novel paradigm to improve the faithfulness of T2I models by fine-tuning models on automatically generated, multi-skill image-text datasets, with skill-specific expert learning and merging. First, SELMA leverages an LLM’s in-context learning capability to generate multiple datasets of text prompts that can teach different skills, and then generates the images with a T2I model based on the prompts. Next, SELMA adapts the T2I model to the new skills by learning multiple single-skill LoRA (low-rank adaptation) experts followed by expert merging. Our independent expert fine-tuning specializes multiple models for different skills, and expert merging helps build a joint multi-skill T2I model that can generate faithful images given diverse text prompts, while mitigating the knowledge conflict from different datasets. We empirically demonstrate that SELMA significantly improves the semantic alignment and text faithfulness of state-of-the-art T2I diffusion models on multiple benchmarks (+2.1% on TIFA and +6.9% on DSG), human preference metrics (PickScore, ImageReward, and HPS), as well as human evaluation. Moreover, fine-tuning with image-text pairs auto-collected via SELMA shows comparable performance to fine-tuning with ground truth data. Lastly, we show that fine-tuning with images from a weaker T2I model can help improve the generation quality of a stronger T2I model, suggesting promising weak-to-strong generalization in T2I models. We provide code in the supplementary materials.

SELMA: Learning and Merging Skill-Specific Text-to-Image Experts with Auto-Generated Data

NeurIPS.cc/2024/Conference

2403.06952

https://huggingface.co/papers/2403.06952

poster

https://openreview.net/forum?id=t8iosEWoyd

@inproceedings{ wen2024stochastic, title={Stochastic contextual bandits with graph feedback: from independence number to {MAS} number}, author={Yuxiao Wen and Yanjun Han and Zhengyuan Zhou}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t8iosEWoyd} }

We consider contextual bandits with graph feedback, a class of interactive learning problems with richer structures than vanilla contextual bandits, where taking an action reveals the rewards for all neighboring actions in the feedback graph under all contexts. Unlike the multi-armed bandits setting where a growing literature has painted a near-complete understanding of graph feedback, much remains unexplored in the contextual bandits counterpart. In this paper, we make inroads into this inquiry by establishing a regret lower bound $\Omega(\sqrt{\beta_M(G) T})$, where $M$ is the number of contexts, $G$ is the feedback graph, and $\beta_M(G)$ is our proposed graph-theoretic quantity that characterizes the fundamental learning limit for this class of problems. Interestingly, $\beta_M(G)$ interpolates between $\alpha(G)$ (the independence number of the graph) and $\mathsf{m}(G)$ (the maximum acyclic subgraph (MAS) number of the graph) as the number of contexts $M$ varies. We also provide algorithms that achieve near-optimal regret for important classes of context sequences and/or feedback graphs, such as transitively closed graphs that find applications in auctions and inventory control. In particular, with many contexts, our results show that the MAS number essentially characterizes the statistical complexity for contextual bandits, as opposed to the independence number in multi-armed bandits.

Stochastic contextual bandits with graph feedback: from independence number to MAS number

NeurIPS.cc/2024/Conference

2402.18591

poster

https://openreview.net/forum?id=t7wvJstsiV

@inproceedings{ zhang2024sled, title={{SLED}: Self Logits Evolution Decoding for Improving Factuality in Large Language Models}, author={Jianyi Zhang and Da-Cheng Juan and Cyrus Rashtchian and Chun-Sung Ferng and Heinrich Jiang and Yiran Chen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t7wvJstsiV} }

Large language models (LLMs) have demonstrated remarkable capabilities, but their outputs can sometimes be unreliable or factually incorrect. To address this, we introduce Self Logits Evolution Decoding (SLED), a novel decoding framework that enhances the truthfulness of LLMs without relying on external knowledge bases or requiring further fine-tuning. From an optimization perspective, our SLED framework leverages the latent knowledge embedded within the LLM by contrasting the output logits from the final layer with those from early layers. It then utilizes an approximate gradient approach to enable latent knowledge to guide the self-refinement of outputs, thereby effectively improving factual accuracy. Extensive experiments have been conducted on established benchmarks across a diverse range of model families (LLaMA 2, LLaMA 3, Gemma) and scales (from 2B to 70B), including more advanced architectural configurations such as the mixture of experts (MoE). Our evaluation spans a wide variety of tasks, including multi-choice, open-generation, and adaptations to chain-of-thought reasoning tasks. The results demonstrate that SLED consistently improves factual accuracy by up to 20\% compared to existing decoding methods while maintaining natural language fluency and negligible latency overhead. Furthermore, it can be flexibly combined with other decoding methods to further enhance their performance.

SLED: Self Logits Evolution Decoding for Improving Factuality in Large Language Models

NeurIPS.cc/2024/Conference

2411.02433

poster

https://openreview.net/forum?id=t7euV5dl5M

@inproceedings{ maus2024approximationaware, title={Approximation-Aware Bayesian Optimization}, author={Natalie Maus and Kyurae Kim and David Eriksson and Geoff Pleiss and John Patrick Cunningham and Jacob R. Gardner}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t7euV5dl5M} }

High-dimensional Bayesian optimization (BO) tasks such as molecular design often require $>10,$$000$ function evaluations before obtaining meaningful results. While methods like sparse variational Gaussian processes (SVGPs) reduce computational requirements in these settings, the underlying approximations result in suboptimal data acquisitions that slow the progress of optimization. In this paper we modify SVGPs to better align with the goals of BO: targeting informed data acquisition over global posterior fidelity. Using the framework of utility-calibrated variational inference (Lacoste–Julien et al., 2011), we unify GP approximation and data acquisition into a joint optimization problem, thereby ensuring optimal decisions under a limited computational budget. Our approach can be used with any decision-theoretic acquisition function and is readily compatible with trust region methods like TuRBO (Eriksson et al., 2019). We derive efficient joint objectives for the expected improvement (EI) and knowledge gradient (KG) acquisition functions in both the standard and batch BO settings. On a variety of recent high dimensional benchmark tasks in control and molecular design, our approach significantly outperforms standard SVGPs and is capable of achieving comparable rewards with up to $10\times$ fewer function evaluations.

Approximation-Aware Bayesian Optimization

NeurIPS.cc/2024/Conference

2406.04308

oral

https://openreview.net/forum?id=t7SGOv5W5z

@inproceedings{ dai2024uqe, title={{UQE}: A Query Engine for Unstructured Databases}, author={Hanjun Dai and Bethany Yixin Wang and Xingchen Wan and Bo Dai and Sherry Yang and Azade Nova and Pengcheng Yin and Phitchaya Mangpo Phothilimthana and Charles Sutton and Dale Schuurmans}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t7SGOv5W5z} }

Analytics on structured data is a mature field with many successful methods. However, most real world data exists in unstructured form, such as images and conversations. We investigate the potential of Large Language Models (LLMs) to enable unstructured data analytics. In particular, we propose a new Universal Query Engine (UQE) that directly interrogates and draws insights from unstructured data collections. This engine accepts queries in a Universal Query Language (UQL), a dialect of SQL that provides full natural language flexibility in specifying conditions and operators. The new engine leverages the ability of LLMs to conduct analysis of unstructured data, while also allowing us to exploit advances in sampling and optimization techniques to achieve efficient and accurate query execution. In addition, we borrow techniques from classical compiler theory to better orchestrate the workflow between sampling methods and foundation model calls. We demonstrate the efficiency of UQE on data analytics across different modalities, including images, dialogs and reviews, across a range of useful query types, including conditional aggregation, semantic retrieval and abstraction aggregation.

UQE: A Query Engine for Unstructured Databases

NeurIPS.cc/2024/Conference

2407.09522

poster

https://openreview.net/forum?id=t4VwoIYBf0

@inproceedings{ luo2024cgail, title={C-{GAIL}: Stabilizing Generative Adversarial Imitation Learning with Control Theory}, author={Tianjiao Luo and Tim Pearce and Huayu Chen and Jianfei Chen and Jun Zhu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t4VwoIYBf0} }

Generative Adversarial Imitation Learning (GAIL) provides a promising approach to training a generative policy to imitate a demonstrator. It uses on-policy Reinforcement Learning (RL) to optimize a reward signal derived from an adversarial discriminator. However, optimizing GAIL is difficult in practise, with the training loss oscillating during training, slowing convergence. This optimization instability can prevent GAIL from finding a good policy, harming its final performance. In this paper, we study GAIL’s optimization from a control-theoretic perspective. We show that GAIL cannot converge to the desired equilibrium. In response, we analyze the training dynamics of GAIL in function space and design a novel controller that not only pushes GAIL to the desired equilibrium but also achieves asymptotic stability in a simplified “one-step” setting. Going from theory to practice, we propose Controlled-GAIL (C-GAIL), which adds a differentiable regularization term on the GAIL objective to stabilize training. Empirically, the C-GAIL regularizer improves the training of various existing GAIL methods, including the popular GAIL-DAC, by speeding up the convergence, reducing the range of oscillation, and matching the expert distribution more closely.

C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control Theory

NeurIPS.cc/2024/Conference

2402.16349

poster

https://openreview.net/forum?id=t3BhmwAzhv

@inproceedings{ huang2024chatscene, title={Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifiers}, author={Haifeng Huang and Yilun Chen and Zehan Wang and Rongjie Huang and Runsen Xu and Tai Wang and Luping Liu and Xize Cheng and Yang Zhao and Jiangmiao Pang and Zhou Zhao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=t3BhmwAzhv} }

Recent advancements in 3D Large Language Models (LLMs) have demonstrated promising capabilities for 3D scene understanding. However, previous methods exhibit deficiencies in general referencing and grounding capabilities for intricate scene comprehension. In this paper, we introduce the use of object identifiers and object-centric representations to interact with scenes at the object level. Specifically, we decompose the input 3D scene into a set of object proposals, each assigned a unique identifier token, which enables efficient object referencing and grounding during user-assistant interactions. Given the scarcity of scene-language data, we model the scene embeddings as a sequence of explicit object-level embeddings, derived from semantic-rich 2D or 3D representations. By employing object identifiers, we transform diverse 3D scene-language tasks into a unified question-answering format, facilitating joint training without the need for additional task-specific heads. With minimal fine-tuning on all downstream tasks, our model significantly outperforms existing methods on benchmarks including ScanRefer, Multi3DRefer, Scan2Cap, ScanQA, and SQA3D.

Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifiers

NeurIPS.cc/2024/Conference

2312.08168

https://huggingface.co/papers/2312.08168

poster

https://openreview.net/forum?id=sy7eSEXdPC

@inproceedings{ estornell2024multillm, title={Multi-{LLM} Debate: Framework, Principals, and Interventions}, author={Andrew Estornell and Yang Liu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sy7eSEXdPC} }

The flexible and generalized nature of large language models has allowed for their application in a wide array of language-based domains. Much like their human contemporaries, these models are capable of engaging in discussions and debates as a means of improving answer quality. We first take a theoretical approach to analyzing debate and provide a framework through which debate can be mathematically examined. Building on this framework, we provide several theoretical results for multi-agent debate. In particular, we demonstrate that similar model capabilities, or similar model responses, can result in static debate dynamics where the debate procedure simply converges to the majority opinion. When this majority opinion is the result of a common misconception (ingrained in the models through shared training data) debate is likely to converge to answers associated with that common misconception. Using insights from our theoretical results we then propose three interventions which improve the efficacy of debate. For each intervention, we provide theoretical results demonstrating how debate is improved. We also demonstrate that these interventions result in better performance on four common benchmark tasks.

Multi-LLM Debate: Framework, Principals, and Interventions

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=sy2SmstDOB

@inproceedings{ zhang2024unifl, title={Uni{FL}: Improve Latent Diffusion Model via Unified Feedback Learning}, author={Jiacheng Zhang and Jie Wu and Yuxi Ren and Xin Xia and Huafeng Kuang and Pan Xie and Jiashi Li and Xuefeng Xiao and Weilin Huang and Shilei Wen and Lean Fu and Guanbin Li}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sy2SmstDOB} }

Latent diffusion models (LDM) have revolutionized text-to-image generation, leading to the proliferation of various advanced models and diverse downstream applications. However, despite these significant advancements, current diffusion models still suffer from several limitations, including inferior visual quality, inadequate aesthetic appeal, and inefficient inference, without a comprehensive solution in sight. To address these challenges, we present **UniFL**, a unified framework that leverages feedback learning to enhance diffusion models comprehensively. UniFL stands out as a universal, effective, and generalizable solution applicable to various diffusion models, such as SD1.5 and SDXL. Notably, UniFL consists of three key components: perceptual feedback learning, which enhances visual quality; decoupled feedback learning, which improves aesthetic appeal; and adversarial feedback learning, which accelerates inference. In-depth experiments and extensive user studies validate the superior performance of our method in enhancing generation quality and inference acceleration. For instance, UniFL surpasses ImageReward by 17\% user preference in terms of generation quality and outperforms LCM and SDXL Turbo by 57\% and 20\% general preference with 4-step inference.

UniFL: Improve Latent Diffusion Model via Unified Feedback Learning

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=swp3lPDmZe

@inproceedings{ gao2024offpolicy, title={Off-Policy Selection for Initiating Human-Centric Experimental Design}, author={Ge Gao and Xi Yang and Qitong Gao and Song Ju and Miroslav Pajic and Min Chi}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=swp3lPDmZe} }

In human-centric applications like healthcare and education, the \textit{heterogeneity} among patients and students necessitates personalized treatments and instructional interventions. While reinforcement learning (RL) has been utilized in those tasks, off-policy selection (OPS) is pivotal to close the loop by offline evaluating and selecting policies without online interactions, yet current OPS methods often overlook the heterogeneity among participants. Our work is centered on resolving a \textit{pivotal challenge} in human-centric systems (HCSs): \textbf{\textit{how to select a policy to deploy when a new participant joining the cohort, without having access to any prior offline data collected over the participant?}} We introduce First-Glance Off-Policy Selection (FPS), a novel approach that systematically addresses participant heterogeneity through sub-group segmentation and tailored OPS criteria to each sub-group. By grouping individuals with similar traits, FPS facilitates personalized policy selection aligned with unique characteristics of each participant or group of participants. FPS is evaluated via two important but challenging applications, intelligent tutoring systems and a healthcare application for sepsis treatment and intervention. FPS presents significant advancement in enhancing learning outcomes of students and in-hospital care outcomes.

Off-Policy Selection for Initiating Human-Centric Experimental Design

NeurIPS.cc/2024/Conference

2410.20017

poster

https://openreview.net/forum?id=suYAAOI5bd

@inproceedings{ yin2024on, title={On the Expressive Power of Tree-Structured Probabilistic Circuits}, author={Lang Yin and Han Zhao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=suYAAOI5bd} }

Probabilistic circuits (PCs) have emerged as a powerful framework compactly representing probability distributions for efficient and exact probabilistic inference. It has been shown that PCs with general directed acyclic graph (DAG) structure can be understood as a mixture of exponentially (in its height) many components, each of which is a product distributions over univariate marginals. However, existing structure learning algorithms for PCs often generate tree-structured circuits, or using tree-structured circuits as intermediate steps to compress them into DAG-structured circuits. This leads to an intriguing question on whether there exists an exponential gap between DAGs and trees for the PC structure. In this paper, we provide a negative answer to this conjecture by proving that, for $n$ variables, there is a quasi-polynomial upper bound $n^{O(\log n)}$ on the size of an equivalent tree computing the same probability distribution. On the other hand, we will also show that given a depth restriction on the tree, there is a super-polynomial separation between tree and DAG-structured PCs. Our work takes an important step towards understanding the expressive power of tree-structured PCs, and our techniques may be of independent interest in the study of structure learning algorithms for PCs.

On the Expressive Power of Tree-Structured Probabilistic Circuits

NeurIPS.cc/2024/Conference

2410.05465

poster

https://openreview.net/forum?id=stY80vVBS8

@inproceedings{ agarwal2024learningaugmented, title={Learning-Augmented Dynamic Submodular Maximization}, author={Arpit Agarwal and Eric Balkanski}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=stY80vVBS8} }

In dynamic submodular maximization, the goal is to maintain a high-value solution over a sequence of element insertions and deletions with a fast update time. Motivated by large-scale applications and the fact that dynamic data often exhibits patterns, we ask the following question: can predictions be used to accelerate the update time of dynamic submodular maximization algorithms? We consider the model for dynamic algorithms with predictions where predictions regarding the insertion and deletion times of elements can be used for preprocessing. Our main result is an algorithm with an $O(\text{poly}(\log \eta, \log w, \log k))$ amortized update time over the sequence of updates that achieves a $1/2 - \epsilon$ approximation for dynamic monotone submodular maximization under a cardinality constraint $k$, where the prediction error $\eta$ is the number of elements that are not inserted and deleted within $w$ time steps of their predicted insertion and deletion times. This amortized update time is independent of the length of the stream and instead depends on the prediction error.

Learning-Augmented Dynamic Submodular Maximization

NeurIPS.cc/2024/Conference

2311.13006

poster

https://openreview.net/forum?id=stXtBqyTWX

@inproceedings{ huang2024toward, title={Toward Efficient Inference for Mixture of Experts}, author={Haiyang Huang and Newsha Ardalani and Anna Sun and Liu Ke and Shruti Bhosale and Hsien-Hsin S. Lee and Carole-Jean Wu and Benjamin Lee}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=stXtBqyTWX} }

Mixture-of-Experts (MoE) models have recently gained steam in achieving the state-of-the-art performance in a wide range of tasks in computer vision and natural language processing. They effectively expand the model capacity while incurring a minimal increase in computation cost during training. However, deploying such models for inference is difficult due to their large model size and complex communication pattern. In this work, we provide a characterization of two MoE workloads, namely Language Modeling (LM) and Machine Translation (MT) and identify their sources of inefficiencies at deployment. We propose three optimization techniques to mitigate sources of inefficiencies, namely (1) Dynamic gating, (2) Expert Buffering, and (3) Expert load balancing. We show that dynamic gating improves maximum throughput by 6.21-11.55$\times$ for LM, 5.75-10.98$\times$ for MT Encoder and 2.58-5.71$\times$ for MT Decoder. It also reduces memory usage by up to 1.36$\times$ for LM and up to 1.1$\times$ for MT. We further propose Expert Buffering, a new caching mechanism that only keeps hot, active experts in GPU memory while buffering the rest in CPU memory. This reduces static memory allocation by 1.47$\times$. Finally, we propose a load balancing methodology that provides additional robustness to the workload. Our code is available at https://github.com/hyhuang00/moe_inference.

Toward Efficient Inference for Mixture of Experts

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=srQxkSPJLW

@inproceedings{ nan2024dimaskdino, title={{DI}-Mask{DINO}: A Joint Object Detection and Instance Segmentation Model}, author={Zhixiong Nan and Xianghong Li and Tao Xiang and Jifeng Dai}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=srQxkSPJLW} }

This paper is motivated by an interesting phenomenon: the performance of object detection lags behind that of instance segmentation (i.e., performance imbalance) when investigating the intermediate results from the beginning transformer decoder layer of MaskDINO (i.e., the SOTA model for joint detection and segmentation). This phenomenon inspires us to think about a question: will the performance imbalance at the beginning layer of transformer decoder constrain the upper bound of the final performance? With this question in mind, we further conduct qualitative and quantitative pre-experiments, which validate the negative impact of detection-segmentation imbalance issue on the model performance. To address this issue, this paper proposes DI-MaskDINO model, the core idea of which is to improve the final performance by alleviating the detection-segmentation imbalance. DI-MaskDINO is implemented by configuring our proposed De-Imbalance (DI) module and Balance-Aware Tokens Optimization (BATO) module to MaskDINO. DI is responsible for generating balance-aware query, and BATO uses the balance-aware query to guide the optimization of the initial feature tokens. The balance-aware query and optimized feature tokens are respectively taken as the Query and Key&Value of transformer decoder to perform joint object detection and instance segmentation. DI-MaskDINO outperforms existing joint object detection and instance segmentation models on COCO and BDD100K benchmarks, achieving +1.2 $AP^{box}$ and +0.9 $AP^{mask}$ improvements compared to SOTA joint detection and segmentation model MaskDINO. In addition, DI-MaskDINO also obtains +1.0 $AP^{box}$ improvement compared to SOTA object detection model DINO and +3.0 $AP^{mask}$ improvement compared to SOTA segmentation model Mask2Former.

DI-MaskDINO: A Joint Object Detection and Instance Segmentation Model

NeurIPS.cc/2024/Conference

2410.16707

poster

https://openreview.net/forum?id=spwE9sLrfg

@inproceedings{ bhatia2024verified, title={Verified Code Transpilation with {LLM}s}, author={Sahil Bhatia and Jie Qiu and Niranjan Hasabnis and Sanjit A. Seshia and Alvin Cheung}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=spwE9sLrfg} }

Domain-specific languages (DSLs) have become integral to various software workflows. Such languages offer domain-specific optimizations and abstractions that improve code readability and maintainability. However, leveraging these languages requires developers to rewrite existing code using the specific DSL's API. While large language models (LLMs) have shown some success in automatic code transpilation, none of them provide any functional correctness guarantees on the rewritten code. Another approach for automating this task is verified lifting, which relies on program synthesis to find programs in the target language that are functionally equivalent to the source language program. While several verified lifting tools have been developed for various application domains, they are specialized for specific source-target languages or require significant expertise in domain knowledge to make the search efficient. In this paper, leveraging recent advances in LLMs, we propose an LLM-based approach (LLMLift) to building verified lifting tools. We use the LLM's capabilities to reason about programs to translate a given program into its corresponding equivalent in the target language. Additionally, we use LLMs to generate proofs for functional equivalence. We develop lifting-based compilers for four DSLs targeting different application domains. Our approach not only outperforms previous symbolic-based tools in number of benchmarks transpiled and transpilation time, but also requires significantly less effort to build.

Verified Code Transpilation with LLMs

NeurIPS.cc/2024/Conference

2406.03003

poster

https://openreview.net/forum?id=sp8wHIsnu9

@inproceedings{ wen2024synthesize, title={Synthesize, Partition, then Adapt: Eliciting Diverse Samples from Foundation Models}, author={Yeming Wen and Swarat Chaudhuri}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sp8wHIsnu9} }

Presenting users with diverse responses from foundation models is crucial for enhancing user experience and accommodating varying preferences. However, generating multiple high-quality and diverse responses without sacrificing accuracy remains a challenge, especially when using greedy sampling. In this work, we propose a novel framework, Synthesize-Partition-Adapt (SPA), that leverages the abundant synthetic data available in many domains to elicit diverse responses from foundation models. By leveraging signal provided by data attribution methods such as influence functions, SPA partitions data into subsets, each targeting unique aspects of the data, and trains multiple model adaptations optimized for these subsets. Experimental results demonstrate the effectiveness of our approach in diversifying foundation model responses while maintaining high quality, showcased through the HumanEval and MBPP tasks in the code generation domain and several tasks in the natural language understanding domain, highlighting its potential to enrich user experience across various applications.

Synthesize, Partition, then Adapt: Eliciting Diverse Samples from Foundation Models

NeurIPS.cc/2024/Conference

2411.06722

poster

https://openreview.net/forum?id=soUXmwL5aK

@inproceedings{ mctavish2024interpretable, title={Interpretable Generalized Additive Models for Datasets with Missing Values}, author={Hayden McTavish and Jon Donnelly and Margo Seltzer and Cynthia Rudin}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=soUXmwL5aK} }

Many important datasets contain samples that are missing one or more feature values. Maintaining the interpretability of machine learning models in the presence of such missing data is challenging. Singly or multiply imputing missing values complicates the model’s mapping from features to labels. On the other hand, reasoning on indicator variables that represent missingness introduces a potentially large number of additional terms, sacrificing sparsity. We solve these problems with M-GAM, a sparse, generalized, additive modeling approach that incorporates missingness indicators and their interaction terms while maintaining sparsity through $\ell_0$ regularization. We show that M-GAM provides similar or superior accuracy to prior methods while significantly improving sparsity relative to either imputation or naïve inclusion of indicator variables.

Interpretable Generalized Additive Models for Datasets with Missing Values

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=snxWD0Q4EI

@inproceedings{ wu2024the, title={The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information}, author={Diyuan Wu and Ionut-Vlad Modoranu and Mher Safaryan and Denis Kuznedelev and Dan Alistarh}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=snxWD0Q4EI} }

The rising footprint of machine learning has led to a focus on imposing model sparsity as a means of reducing computational and memory costs. For deep neural networks (DNNs), the state-of-the-art accuracy-vs-sparsity is achieved by heuristics inspired by the classical Optimal Brain Surgeon (OBS) framework [LeCun et al., 1989, Hassibi and Stork, 1992, Hassibi et al., 1993], which leverages loss curvature information to make better pruning decisions. Yet, these results still lack a solid theoretical understanding, and it is unclear whether they can be improved by leveraging connections to the wealth of work on sparse recovery algorithms. In this paper, we draw new connections between these two areas and present new sparse recovery algorithms inspired by the OBS framework that come with theoretical guarantees under reasonable assumptions and have strong practical performance. Specifically, our work starts from the observation that we can leverage curvature information in OBS-like fashion upon the projection step of classic iterative sparse recovery algorithms such as IHT. We show for the first time that this leads both to improved convergence bounds in well-behaved settings and to stronger practical convergence. Furthermore, we present extensions of this approach to training accurate sparse DNNs, and validate it experimentally at scale.

The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information

NeurIPS.cc/2024/Conference

2408.17163

poster

https://openreview.net/forum?id=sntv8Ac3U2

@inproceedings{ sridhar2024adapting, title={Adapting Diffusion Models for Improved Prompt Compliance and Controllable Image Synthesis}, author={Deepak Sridhar and Abhishek Peri and Rohith Reddy Rachala and Nuno Vasconcelos}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sntv8Ac3U2} }

Recent advances in generative modeling with diffusion processes (DPs) enabled breakthroughs in image synthesis. Despite impressive image quality, these models have various prompt compliance problems, including low recall in generating multiple objects, difficulty in generating text in images, and meeting constraints like object locations and pose. For fine-grained editing and manipulation, they also require fine-grained semantic or instance maps that are tedious to produce manually. While prompt compliance can be enhanced by addition of loss functions at inference, this is time consuming and does not scale to complex scenes. To overcome these limitations, this work introduces a new family of $\textit{Factor Graph Diffusion Models}$ (FG-DMs) that models the joint distribution of images and conditioning variables, such as semantic, sketch, depth or normal maps via a factor graph decomposition. This joint structure has several advantages, including support for efficient sampling based prompt compliance schemes, which produce images of high object recall, semi-automated fine-grained editing, explainability at intermediate levels, ability to produce labeled datasets for the training of downstream models such as segmentation or depth, training with missing data, and continual learning where new conditioning variables can be added with minimal or no modifications to the existing structure. We propose an implementation of FG-DMs by adapting a pre-trained Stable Diffusion (SD) model to implement all FG-DM factors, using only COCO dataset, and show that it is effective in generating images with 15\% higher recall than SD while retaining its generalization ability. We introduce an attention distillation loss that encourages consistency among the attention maps of all factors, improving the fidelity of the generated conditions and image. We also show that training FG-DMs from scratch on MM-CelebA-HQ, Cityscapes, ADE20K, and COCO produce images of high quality (FID) and diversity (LPIPS).

Adapting Diffusion Models for Improved Prompt Compliance and Controllable Image Synthesis

NeurIPS.cc/2024/Conference

2410.21638

poster

https://openreview.net/forum?id=sn3UrYRItk

@inproceedings{ hayou2024the, title={The Impact of Initialization on Lo{RA} Finetuning Dynamics}, author={Soufiane Hayou and Nikhil Ghosh and Bin Yu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sn3UrYRItk} }

In this paper, we study the role of initialization in Low Rank Adaptation (LoRA) as originally introduced in Hu et al. (2021). Essentially, to start from the pretrained model, one can either initialize $B$ to zero and $A$ to random, or vice-versa. In both cases, the product $BA$ is equal to zero at initialization, which makes finetuning starts from the pretrained model. These two initialization schemes are seemingly similar. They should in-principle yield the same performance and share the same optimal learning rate. We demonstrate that this is an *incorrect intuition* and that the first scheme (of initializing $B$ to zero and $A$ to random) on average in our experiments yields better performance compared to the other scheme. Our theoretical analysis shows that the reason behind this might be that the first initialization allows the use of larger learning rates (without causing output instability) compared to the second initialization, resulting in more efficient learning of the first scheme. We validate our results with extensive experiments on LLMs.

The Impact of Initialization on LoRA Finetuning Dynamics

NeurIPS.cc/2024/Conference

2406.08447

poster

https://openreview.net/forum?id=sks7x4I8Bh

@inproceedings{ foster2024online, title={Online Estimation via Offline Estimation: An Information-Theoretic Framework}, author={Dylan J Foster and Yanjun Han and Jian Qian and Alexander Rakhlin}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sks7x4I8Bh} }

The classical theory of statistical estimation aims to estimate a parameter of interest under data generated from a fixed design (''offline estimation''), while the contemporary theory of online learning provides algorithms for estimation under adaptively chosen covariates (''online estimation''). Motivated by connections between estimation and interactive decision making, we ask: is it possible to convert offline estimation algorithms into online estimation algorithms in a black-box fashion? We investigate this question from an information-theoretic perspective by introducing a new framework, Oracle-Efficient Online Estimation (OEOE), where the learner can only interact with the data stream indirectly through a sequence of offline estimators produced by a black-box algorithm operating on the stream. Our main results settle the statistical and computational complexity of online estimation in this framework. $\bullet$ Statistical complexity. We show that information-theoretically, there exist algorithms that achieve near-optimal online estimation error via black-box offline estimation oracles, and give a nearly-tight characterization for minimax rates in the OEOE framework. $\bullet$ Computational complexity. We show that the guarantees above cannot be achieved in a computationally efficient fashion in general, but give a refined characterization for the special case of conditional density estimation: computationally efficient online estimation via black-box offline estimation is possible whenever it is possible via unrestricted algorithms. Finally, we apply our results to give offline oracle-efficient algorithms for interactive decision making.

Online Estimation via Offline Estimation: An Information-Theoretic Framework

NeurIPS.cc/2024/Conference

2404.10122

poster

https://openreview.net/forum?id=skeopn3q5Y

@inproceedings{ lyu2024sfpuel, title={Sf{PUEL}: Shape from Polarization under Unknown Environment Light}, author={Youwei Lyu and Heng Guo and Kailong Zhang and Si Li and Boxin Shi}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=skeopn3q5Y} }

Shape from polarization (SfP) benefits from advancements like polarization cameras for single-shot normal estimation, but its performance heavily relies on light conditions. This paper proposes SfPUEL, an end-to-end SfP method to jointly estimate surface normal and material under unknown environment light. To handle this challenging light condition, we design a transformer-based framework for enhancing the perception of global context features. We further propose to integrate photometric stereo (PS) priors from pretrained models to enrich extracted features for high-quality normal predictions. As metallic and dielectric materials exhibit different BRDFs, SfPUEL additionally predicts dielectric and metallic material segmentation to further boost performance. Experimental results on synthetic and our collected real-world dataset demonstrate that SfPUEL significantly outperforms existing SfP and single-shot normal estimation methods. The code and dataset is available at https://github.com/YouweiLyu/SfPUEL.

SfPUEL: Shape from Polarization under Unknown Environment Light

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=siPdcro6uD

@inproceedings{ xiao2024oneref, title={OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling}, author={Linhui Xiao and Xiaoshan Yang and Fang Peng and Yaowei Wang and Changsheng Xu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=siPdcro6uD} }

Constrained by the separate encoding of vision and language, existing grounding and referring segmentation works heavily rely on bulky Transformer-based fusion en-/decoders and a variety of early-stage interaction technologies. Simultaneously, the current mask visual language modeling (MVLM) fails to capture the nuanced referential relationship between image-text in referring tasks. In this paper, we propose **OneRef**, a minimalist referring framework built on the modality-shared one-tower transformer that unifies the visual and linguistic feature spaces. To modeling the referential relationship, we introduce a novel MVLM paradigm called Mask Referring Modeling (**MRefM**), which encompasses both referring-aware mask image modeling and referring-aware mask language modeling. Both modules not only reconstruct modality-related content but also cross-modal referring content. Within MRefM, we propose a referring-aware dynamic image masking strategy that is aware of the referred region rather than relying on fixed ratios or generic random masking schemes. By leveraging the unified visual language feature space and incorporating MRefM's ability to model the referential relations, our approach enables direct regression of the referring results without resorting to various complex techniques. Our method consistently surpasses existing approaches and achieves SoTA performance on both grounding and segmentation tasks, providing valuable insights for future research. Our code and models are available at https://github.com/linhuixiao/OneRef.

OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=shYQXpnBLB

@inproceedings{ zhou2024association, title={Association of Objects May Engender Stereotypes: Mitigating Association-Engendered Stereotypes in Text-to-Image Generation}, author={Junlei Zhou and Jiashi Gao and Xiangyu Zhao and Xin Yao and Xuetao Wei}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=shYQXpnBLB} }

Text-to-Image (T2I) has witnessed significant advancements, demonstrating superior performance for various generative tasks. However, the presence of stereotypes in T2I introduces harmful biases that require urgent attention as the T2I technology becomes more prominent. Previous work for stereotype mitigation mainly concentrated on mitigating stereotypes engendered with individual objects within images, which failed to address stereotypes engendered by the association of multiple objects, referred to as *Association-Engendered Stereotypes*. For example, mentioning ''black people'' and ''houses'' separately in prompts may not exhibit stereotypes. Nevertheless, when these two objects are associated in prompts, the association of ''black people'' with ''poorer houses'' becomes more pronounced. To tackle this issue, we propose a novel framework, MAS, to Mitigate Association-engendered Stereotypes. This framework models the stereotype problem as a probability distribution alignment problem, aiming to align the stereotype probability distribution of the generated image with the stereotype-free distribution. The MAS framework primarily consists of the *Prompt-Image-Stereotype CLIP* (*PIS CLIP*) and *Sensitive Transformer*. The *PIS CLIP* learns the association between prompts, images, and stereotypes, which can establish the mapping of prompts to stereotypes. The *Sensitive Transformer* produces the sensitive constraints, which guide the stereotyped image distribution to align with the stereotype-free probability distribution. Moreover, recognizing that existing metrics are insufficient for accurately evaluating association-engendered stereotypes, we propose a novel metric, *Stereotype-Distribution-Total-Variation*(*SDTV*), to evaluate stereotypes in T2I. Comprehensive experiments demonstrate that our framework effectively mitigates association-engendered stereotypes.

Association of Objects May Engender Stereotypes: Mitigating Association-Engendered Stereotypes in Text-to-Image Generation

NeurIPS.cc/2024/Conference

oral

https://openreview.net/forum?id=sgVOjDqUMT

@inproceedings{ liu2024minicache, title={MiniCache: {KV} Cache Compression in Depth Dimension for Large Language Models}, author={Akide Liu and Jing Liu and Zizheng Pan and Yefei He and Gholamreza Haffari and Bohan Zhuang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sgVOjDqUMT} }

A critical approach for efficiently deploying computationally demanding large language models (LLMs) is Key-Value (KV) caching. The KV cache stores key-value states of previously generated tokens, significantly reducing the need for repetitive computations and thereby lowering latency in autoregressive generation. However, the size of the KV cache grows linearly with sequence length, posing challenges for applications requiring long context input and extensive sequence generation. In this paper, we present a simple yet effective approach, called MiniCache, to compress the KV cache across layers from a novel depth perspective, significantly reducing the memory footprint for LLM inference. Our approach is based on the observation that KV cache states exhibit high similarity between the adjacent layers in the middle-to-deep portion of LLMs. To facilitate merging, we propose disentangling the states into the magnitude and direction components, interpolating the directions of the state vectors while preserving their lengths unchanged. Furthermore, we introduce a token retention strategy to keep highly distinct state pairs unmerged, thus preserving the information with minimal additional storage overhead. Our MiniCache is training-free and general, complementing existing KV cache compression strategies, such as quantization and sparsity. We conduct a comprehensive evaluation of MiniCache utilizing various models including LLaMA-2, LLaMA-3, Phi-3, Mistral, and Mixtral across multiple benchmarks, demonstrating its exceptional performance in achieving superior compression ratios and high throughput. On the ShareGPT dataset, LLaMA-2-7B with cross-layer merging achieves a compression ratio of $1.53\times$. Additionally, since MiniCache is orthogonal to existing quantization techniques, it can achieve a compression ratio of up to $5.02\times$ when combined with the 4-bit quantization technique, enhancing inference throughput by approximately $5\times$ and reducing the memory footprint by $41\%$ compared to the FP16 full cache baseline, all while maintaining near-lossless performance. Project is available at https://minicache.vmv.re .

MiniCache: KV Cache Compression in Depth Dimension for Large Language Models

NeurIPS.cc/2024/Conference

2405.14366

https://huggingface.co/papers/2405.14366

poster

https://openreview.net/forum?id=sfPxUqzdPI

@inproceedings{ ren2024multiscale, title={Multi-scale Consistency for Robust 3D Registration via Hierarchical Sinkhorn Tree}, author={Chengwei Ren and Yifan Feng and Weixiang Zhang and Xiao-Ping Zhang and Yue Gao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sfPxUqzdPI} }

We study the problem of retrieving accurate correspondence through multi-scale consistency (MSC) for robust point cloud registration. Existing works in a coarse-to-fine manner either suffer from severe noisy correspondences caused by unreliable coarse matching or struggle to form outlier-free coarse-level correspondence sets. To tackle this, we present Hierarchical Sinkhorn Tree (HST), a pruned tree structure designed to hierarchically measure the local consistency of each coarse correspondence across multiple feature scales, thereby filtering out the local dissimilar ones. In this way, we convert the modeling of MSC for each correspondence into a BFS traversal with pruning of a K-ary tree rooted at the superpoint, with its K nearest neighbors in the feature pyramid serving as child nodes. To achieve efficient pruning and accurate vicinity characterization, we further propose a novel overlap-aware Sinkhorn Distance, which retains only the most likely overlapping points for local measurement and next level exploration. The modeling process essentially involves traversing a pair of HSTs synchronously and aggregating the consistency measures of corresponding tree nodes. Extensive experiments demonstrate HST consistently outperforms the state-of-the-art methods on both indoor and outdoor benchmarks.

Multi-scale Consistency for Robust 3D Registration via Hierarchical Sinkhorn Tree

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=seYXqfGT0q

@inproceedings{ zheng2024prototypical, title={Prototypical Hash Encoding for On-the-Fly Fine-Grained Category Discovery}, author={Haiyang Zheng and Nan Pu and Wenjing Li and Nicu Sebe and Zhun Zhong}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=seYXqfGT0q} }

In this paper, we study a practical yet challenging task, On-the-fly Category Discovery (OCD), aiming to online discover the newly-coming stream data that belong to both known and unknown classes, by leveraging only known category knowledge contained in labeled data. Previous OCD methods employ the hash-based technique to represent old/new categories by hash codes for instance-wise inference. However, directly mapping features into low-dimensional hash space not only inevitably damages the ability to distinguish classes and but also causes ``high sensitivity'' issue, especially for fine-grained classes, leading to inferior performance. To address these drawbacks, we propose a novel Prototypical Hash Encoding (PHE) framework consisting of Category-aware Prototype Generation (CPG) and Discriminative Category Encoding (DCE) to mitigate the sensitivity of hash code while preserving rich discriminative information contained in high-dimension feature space, in a two-stage projection fashion. CPG enables the model to fully capture the intra-category diversity by representing each category with multiple prototypes. DCE boosts the discrimination ability of hash code with the guidance of the generated category prototypes and the constraint of minimum separation distance. By jointly optimizing CPG and DCE, we demonstrate that these two components are mutually beneficial towards an effective OCD. Extensive experiments show the significant superiority of our PHE over previous methods, e.g. obtaining an improvement of +5.3% in ALL ACC averaged on all datasets. Moreover, due to the nature of the interpretable prototypes, we visually analyze the underlying mechanism of how PHE helps group certain samples into either known or unknown categories. Code is available at https://github.com/HaiyangZheng/PHE.

Prototypical Hash Encoding for On-the-Fly Fine-Grained Category Discovery

NeurIPS.cc/2024/Conference

2410.19213

poster

https://openreview.net/forum?id=seAuMedrm5

@inproceedings{ stooke2024alignerencoders, title={Aligner-Encoders: Self-Attention Transformers Can Be Self-Transducers}, author={Adam Stooke and Rohit Prabhavalkar and Khe Chai Sim and Pedro J Moreno Mengibar}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=seAuMedrm5} }

Modern systems for automatic speech recognition, including the RNN-Transducer and Attention-based Encoder-Decoder (AED), are designed so that the encoder is not required to alter the time-position of information from the audio sequence into the embedding; alignment to the final text output is processed during decoding. We discover that the transformer-based encoder adopted in recent years is actually capable of performing the alignment internally during the forward pass, prior to decoding. This new phenomenon enables a simpler and more efficient model, the ''Aligner-Encoder''. To train it, we discard the dynamic programming of RNN-T in favor of the frame-wise cross-entropy loss of AED, while the decoder employs the lighter text-only recurrence of RNN-T without learned cross-attention---it simply scans embedding frames in order from the beginning, producing one token each until predicting the end-of-message. We conduct experiments demonstrating performance remarkably close to the state of the art, including a special inference configuration enabling long-form recognition. In a representative comparison, we measure the total inference time for our model to be 2x faster than RNN-T and 16x faster than AED. Lastly, we find that the audio-text alignment is clearly visible in the self-attention weights of a certain layer, which could be said to perform ''self-transduction''.

Aligner-Encoders: Self-Attention Transformers Can Be Self-Transducers

NeurIPS.cc/2024/Conference

oral

https://openreview.net/forum?id=scw6Et4pEr

@inproceedings{ ma2024deeplag, title={DeepLag: Discovering Deep Lagrangian Dynamics for Intuitive Fluid Prediction}, author={Qilong Ma and Haixu Wu and Lanxiang Xing and Shangchen Miao and Mingsheng Long}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=scw6Et4pEr} }

Accurately predicting the future fluid is vital to extensive areas such as meteorology, oceanology, and aerodynamics. However, since the fluid is usually observed from the Eulerian perspective, its moving and intricate dynamics are seriously obscured and confounded in static grids, bringing thorny challenges to the prediction. This paper introduces a new Lagrangian-Eulerian combined paradigm to tackle the tanglesome fluid dynamics. Instead of solely predicting the future based on Eulerian observations, we propose DeepLag to discover hidden Lagrangian dynamics within the fluid by tracking the movements of adaptively sampled key particles. Further, DeepLag presents a new paradigm for fluid prediction, where the Lagrangian movement of the tracked particles is inferred from Eulerian observations, and their accumulated Lagrangian dynamics information is incorporated into global Eulerian evolving features to guide future prediction respectively. Tracking key particles not only provides a transparent and interpretable clue for fluid dynamics but also makes our model free from modeling complex correlations among massive grids for better efficiency. Experimentally, DeepLag excels in three challenging fluid prediction tasks covering 2D and 3D, simulated and real-world fluids. Code is available at this repository: https://github.com/thuml/DeepLag.

DeepLag: Discovering Deep Lagrangian Dynamics for Intuitive Fluid Prediction

NeurIPS.cc/2024/Conference

2402.02425

https://huggingface.co/papers/2402.02425

poster

https://openreview.net/forum?id=sbsaRj475E

@inproceedings{ haoweiz2024dipgo, title={DiP-{GO}: A Diffusion Pruner via Few-step Gradient Optimization}, author={haoweiz and Dehua Tang and Ji Liu and Mingjie Lu and Jintu Zheng and Jinzhang Peng and Dong Li and Yu Wang and Fan Jiang and Lu Tian and Spandan Tiwari and Ashish Sirasao and Jun-Hai Yong and Bin Wang and Emad Barsoum}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sbsaRj475E} }

Diffusion models have achieved remarkable progress in the field of image generation due to their outstanding capabilities. However, these models require substantial computing resources because of the multi-step denoising process during inference. While traditional pruning methods have been employed to optimize these models, the retraining process necessitates large-scale training datasets and extensive computational costs to maintain generalization ability, making it neither convenient nor efficient. Recent studies attempt to utilize the similarity of features across adjacent denoising stages to reduce computational costs through simple and static strategies. However, these strategies cannot fully harness the potential of the similar feature patterns across adjacent timesteps. In this work, we propose a novel pruning method that derives an efficient diffusion model via a more intelligent and differentiable pruner. At the core of our approach is casting the model pruning process into a SubNet search process. Specifically, we first introduce a SuperNet based on standard diffusion via adding some backup connections built upon the similar features. We then construct a plugin pruner network and design optimization losses to identify redundant computation. Finally, our method can identify an optimal SubNet through few-step gradient optimization and a simple post-processing procedure. We conduct extensive experiments on various diffusion models including Stable Diffusion series and DiTs. Our DiP-GO approach achieves 4.4 x speedup for SD-1.5 without any loss of accuracy, significantly outperforming the previous state-of-the-art methods.

DiP-GO: A Diffusion Pruner via Few-step Gradient Optimization

NeurIPS.cc/2024/Conference

2410.16942

poster

https://openreview.net/forum?id=satH8Evs2y

@inproceedings{ liu2024beware, title={Beware of Road Markings: A New Adversarial Patch Attack to Monocular Depth Estimation}, author={Hangcheng Liu and Zhenhu Wu and Hao Wang and XINGSHUO HAN and Shangwei Guo and Tao Xiang and Tianwei Zhang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=satH8Evs2y} }

Monocular Depth Estimation (MDE) enables the prediction of scene depths from a single RGB image, having been widely integrated into production-grade autonomous driving systems, e.g., Tesla Autopilot. Current adversarial attacks to MDE models focus on attaching an optimized adversarial patch to a designated obstacle. Although effective, this approach presents two inherent limitations: its reliance on specific obstacles and its limited malicious impact. In contrast, we propose a pioneering attack to MDE models that \textit{decouples obstacles from patches physically and deploys optimized patches on roads}, thereby extending the attack scope to arbitrary traffic participants. This approach is inspired by our groundbreaking discovery: \textit{various MDE models with different architectures, trained for autonomous driving, heavily rely on road regions} when predicting depths for different obstacles. Based on this discovery, we design the Adversarial Road Marking (AdvRM) attack, which camouflages patches as ordinary road markings and deploys them on roads, thereby posing a continuous threat within the environment. Experimental results from both dataset simulations and real-world scenarios demonstrate that AdvRM is effective, stealthy, and robust against various MDE models, achieving about 1.507 of Mean Relative Shift Ratio (MRSR) over 8 MDE models. The code is available at \url{https://github.com/a-c-a-c/AdvRM.git}

Beware of Road Markings: A New Adversarial Patch Attack to Monocular Depth Estimation

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=sZ7jj9kqAy

@inproceedings{ stefa{\'n}ski2024soi, title={{SOI}: Scaling Down Computational Complexity by Estimating Partial States of the Model}, author={Grzegorz Stefa{\'n}ski and Pawe{\l} Daniluk and Artur Szumaczuk and Jakub Tkaczuk}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sZ7jj9kqAy} }

Consumer electronics used to follow the miniaturization trend described by Moore’s Law. Despite increased processing power in Microcontroller Units (MCUs), MCUs used in the smallest appliances are still not capable of running even moderately big, state-of-the-art artificial neural networks (ANNs) especially in time-sensitive scenarios. In this work, we present a novel method called Scattered Online Inference (SOI) that aims to reduce the computational complexity of ANNs. SOI leverages the continuity and seasonality of time-series data and model predictions, enabling extrapolation for processing speed improvements, particularly in deeper layers. By applying compression, SOI generates more general inner partial states of ANN, allowing skipping full model recalculation at each inference.

SOI: Scaling Down Computational Complexity by Estimating Partial States of the Model

NeurIPS.cc/2024/Conference

2410.03813

poster

https://openreview.net/forum?id=sVZBJoxwk9

@inproceedings{ liu2024generalized, title={Generalized Eigenvalue Problems with Generative Priors}, author={Zhaoqiang Liu and Wen Li and Junren Chen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sVZBJoxwk9} }

Generalized eigenvalue problems (GEPs) find applications in various fields of science and engineering. For example, principal component analysis, Fisher's discriminant analysis, and canonical correlation analysis are specific instances of GEPs and are widely used in statistical data processing. In this work, we study GEPs under generative priors, assuming that the underlying leading generalized eigenvector lies within the range of a Lipschitz continuous generative model. Under appropriate conditions, we show that any optimal solution to the corresponding optimization problems attains the optimal statistical rate. Moreover, from a computational perspective, we propose an iterative algorithm called the Projected Rayleigh Flow Method (PRFM) to approximate the optimal solution. We theoretically demonstrate that under suitable assumptions, PRFM converges linearly to an estimated vector that achieves the optimal statistical rate. Numerical results are provided to demonstrate the effectiveness of the proposed method.

Generalized Eigenvalue Problems with Generative Priors

NeurIPS.cc/2024/Conference

2411.01326

poster

https://openreview.net/forum?id=sRSjr9SDKR

@inproceedings{ mikkola2024preferential, title={Preferential Normalizing Flows}, author={Petrus Mikkola and Luigi Acerbi and Arto Klami}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sRSjr9SDKR} }

Eliciting a high-dimensional probability distribution from an expert via noisy judgments is notoriously challenging, yet useful for many applications, such as prior elicitation and reward modeling. We introduce a method for eliciting the expert's belief density as a normalizing flow based solely on preferential questions such as comparing or ranking alternatives. This allows eliciting in principle arbitrarily flexible densities, but flow estimation is susceptible to the challenge of collapsing or diverging probability mass that makes it difficult in practice. We tackle this problem by introducing a novel functional prior for the flow, motivated by a decision-theoretic argument, and show empirically that the belief density can be inferred as the function-space maximum a posteriori estimate. We demonstrate our method by eliciting multivariate belief densities of simulated experts, including the prior belief of a general-purpose large language model over a real-world dataset.

Preferential Normalizing Flows

NeurIPS.cc/2024/Conference

2410.08710

poster

https://openreview.net/forum?id=sRILMnkkQd

@inproceedings{ lin2024unigad, title={Uni{GAD}: Unifying Multi-level Graph Anomaly Detection}, author={Yiqing Lin and Jianheng Tang and Chenyi Zi and H. Vicky Zhao and Yuan Yao and Jia Li}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sRILMnkkQd} }

Graph Anomaly Detection (GAD) aims to identify uncommon, deviated, or suspicious objects within graph-structured data. Existing methods generally focus on a single graph object type (node, edge, graph, etc.) and often overlook the inherent connections among different object types of graph anomalies. For instance, a money laundering transaction might involve an abnormal account and the broader community it interacts with. To address this, we present UniGAD, the first unified framework for detecting anomalies at node, edge, and graph levels jointly. Specifically, we develop the Maximum Rayleigh Quotient Subgraph Sampler (MRQSampler) that unifies multi-level formats by transferring objects at each level into graph-level tasks on subgraphs. We theoretically prove that MRQSampler maximizes the accumulated spectral energy of subgraphs (i.e., the Rayleigh quotient) to preserve the most significant anomaly information. To further unify multi-level training, we introduce a novel GraphStitch Network to integrate information across different levels, adjust the amount of sharing required at each level, and harmonize conflicting training goals. Comprehensive experiments show that UniGAD outperforms both existing GAD methods specialized for a single task and graph prompt-based approaches for multiple tasks, while also providing robust zero-shot task transferability.

UniGAD: Unifying Multi-level Graph Anomaly Detection

NeurIPS.cc/2024/Conference

2411.06427

poster

https://openreview.net/forum?id=sQApQMBqiP

@inproceedings{ wynn2024learning, title={Learning Human-like Representations to Enable Learning Human Values}, author={Andrea Wynn and Ilia Sucholutsky and Thomas L. Griffiths}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sQApQMBqiP} }

How can we build AI systems that can learn any set of individual human values both quickly and safely, avoiding causing harm or violating societal standards for acceptable behavior during the learning process? We explore the effects of representational alignment between humans and AI agents on learning human values. Making AI systems learn human-like representations of the world has many known benefits, including improving generalization, robustness to domain shifts, and few-shot learning performance. We demonstrate that this kind of representational alignment can also support safely learning and exploring human values in the context of personalization. We begin with a theoretical prediction, show that it applies to learning human morality judgments, then show that our results generalize to ten different aspects of human values -- including ethics, honesty, and fairness -- training AI agents on each set of values in a multi-armed bandit setting, where rewards reflect human value judgments over the chosen action. Using a set of textual action descriptions, we collect value judgments from humans, as well as similarity judgments from both humans and multiple language models, and demonstrate that representational alignment enables both safe exploration and improved generalization when learning human values.

Learning Human-like Representations to Enable Learning Human Values

NeurIPS.cc/2024/Conference

2312.14106

poster

https://openreview.net/forum?id=sOhFyFFnxT

@inproceedings{ bond2024exploring, title={Exploring the Precise Dynamics of Single-Layer {GAN} Models: Leveraging Multi-Feature Discriminators for High-Dimensional Subspace Learning}, author={Andrew Bond and Zafer Dogan}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sOhFyFFnxT} }

Subspace learning is a critical endeavor in contemporary machine learning, particularly given the vast dimensions of modern datasets. In this study, we delve into the training dynamics of a single-layer GAN model from the perspective of subspace learning, framing these GANs as a novel approach to this fundamental task. Through a rigorous scaling limit analysis, we offer insights into the behavior of this model. Extending beyond prior research that primarily focused on sequential feature learning, we investigate the non-sequential scenario, emphasizing the pivotal role of inter-feature interactions in expediting training and enhancing performance, particularly with an uninformed initialization strategy. Our investigation encompasses both synthetic and real-world datasets, such as MNIST and Olivetti Faces, demonstrating the robustness and applicability of our findings to practical scenarios. By bridging our analysis to the realm of subspace learning, we systematically compare the efficacy of GAN-based methods against conventional approaches, both theoretically and empirically. Notably, our results unveil that while all methodologies successfully capture the underlying subspace, GANs exhibit a remarkable capability to acquire a more informative basis, owing to their intrinsic ability to generate new data samples. This elucidates the unique advantage of GAN-based approaches in subspace learning tasks.

Exploring the Precise Dynamics of Single-Layer GAN Models: Leveraging Multi-Feature Discriminators for High-Dimensional Subspace Learning

NeurIPS.cc/2024/Conference

2411.00498

poster

https://openreview.net/forum?id=sNz7tptCH6

@inproceedings{ ming2024boosting, title={Boosting the Transferability of Adversarial Attack on Vision Transformer with Adaptive Token Tuning}, author={Di Ming and Peng Ren and Yunlong Wang and Xin Feng}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sNz7tptCH6} }

Vision transformers (ViTs) perform exceptionally well in various computer vision tasks but remain vulnerable to adversarial attacks. Recent studies have shown that the transferability of adversarial examples exists for CNNs, and the same holds true for ViTs. However, existing ViT attacks aggressively regularize the largest token gradients to exact zero within each layer of the surrogate model, overlooking the interactions between layers, which limits their transferability in attacking black-box models. Therefore, in this paper, we focus on boosting the transferability of adversarial attacks on ViTs through adaptive token tuning (ATT). Specifically, we propose three optimization strategies: an adaptive gradient re-scaling strategy to reduce the overall variance of token gradients, a self-paced patch out strategy to enhance the diversity of input tokens, and a hybrid token gradient truncation strategy to weaken the effectiveness of attention mechanism. We demonstrate that scaling correction of gradient changes using gradient variance across different layers can produce highly transferable adversarial examples. In addition, introducing attentional truncation can mitigate the overfitting over complex interactions between tokens in deep ViT layers to further improve the transferability. On the other hand, using feature importance as a guidance to discard a subset of perturbation patches in each iteration, along with combining self-paced learning and progressively more sampled attacks, significantly enhances the transferability over attacks that use all perturbation patches. Extensive experiments conducted on ViTs, undefended CNNs, and defended CNNs validate the superiority of our proposed ATT attack method. On average, our approach improves the attack performance by 10.1% compared to state-of-the-art transfer-based attacks. Notably, we achieve the best attack performance with an average of 58.3% on three defended CNNs. Code is available at https://github.com/MisterRpeng/ATT.

Boosting the Transferability of Adversarial Attack on Vision Transformer with Adaptive Token Tuning

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=sKEhebkEdz

@inproceedings{ zhang2024semisupervised, title={Semi-supervised Knowledge Transfer Across Multi-omic Single-cell Data}, author={Fan Zhang and Tianyu Liu and Zihao Chen and Xiaojiang Peng and Chong Chen and Xian-Sheng Hua and Xiao Luo and Hongyu Zhao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sKEhebkEdz} }

Knowledge transfer between multi-omic single-cell data aims to effectively transfer cell types from scRNA-seq data to unannotated scATAC-seq data. Several approaches aim to reduce the heterogeneity of multi-omic data while maintaining the discriminability of cell types with extensive annotated data. However, in reality, the cost of collecting both a large amount of labeled scRNA-seq data and scATAC-seq data is expensive. Therefore, this paper explores a practical yet underexplored problem of knowledge transfer across multi-omic single-cell data under cell type scarcity. To address this problem, we propose a semi-supervised knowledge transfer framework named Dual label scArcity elimiNation with Cross-omic multi-samplE Mixup (DANCE). To overcome the label scarcity in scRNA-seq data, we generate pseudo-labels based on optimal transport and merge them into the labeled scRNA-seq data. Moreover, we adopt a divide-and-conquer strategy which divides the scATAC-seq data into source-like and target-specific data. For source-like samples, we employ consistency regularization with random perturbations while for target-specific samples, we select a few candidate labels and progressively eliminate incorrect cell types from the label set for additional supervision. Next, we generate virtual scRNA-seq samples with multi-sample Mixup based on the class-wise similarity to reduce cell heterogeneity. Extensive experiments on many benchmark datasets suggest the superiority of our DANCE over a series of state-of-the-art methods.

Semi-supervised Knowledge Transfer Across Multi-omic Single-cell Data

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=sKCKPr8cRL

@inproceedings{ tao2024scaling, title={Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies}, author={Chaofan Tao and Qian Liu and Longxu Dou and Niklas Muennighoff and Zhongwei Wan and Ping Luo and Min Lin and Ngai Wong}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sKCKPr8cRL} }

Research on scaling large language models (LLMs) has primarily focused on model parameters and training data size, overlooking the role of vocabulary size. We investigate how vocabulary size impacts LLM scaling laws by training models ranging from 33M to 3B parameters on up to 500B characters with various vocabulary configurations. We propose three complementary approaches for predicting the compute-optimal vocabulary size: IsoFLOPs analysis, derivative estimation, and parametric fit of the loss function. Our approaches converge on the conclusion that the optimal vocabulary size depends on the compute budget, with larger models requiring larger vocabularies. Most LLMs, however, use insufficient vocabulary sizes. For example, we predict that the optimal vocabulary size of Llama2-70B should have been at least 216K, 7 times larger than its vocabulary of 32K. We validate our predictions empirically by training models with 3B parameters across different FLOPs budgets. Adopting our predicted optimal vocabulary size consistently improves downstream performance over commonly used vocabulary sizes. By increasing the vocabulary size from the conventional 32K to 43K, we improve performance on ARC-Challenge from 29.1 to 32.0 with the same 2.3e21 FLOPs. Our work highlights the importance of jointly considering tokenization and model scaling for efficient pre-training. The code and demo are available at https://github.com/sail-sg/scaling-with-vocab and https://hf.co/spaces/sail/scaling-with-vocab-demo.

Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies

NeurIPS.cc/2024/Conference

2407.13623

https://huggingface.co/papers/2407.13623

poster

https://openreview.net/forum?id=sIsbOkQmBL

@inproceedings{ li2024culturellm, title={Culture{LLM}: Incorporating Cultural Differences into Large Language Models}, author={CHENG LI and Mengzhuo Chen and Jindong Wang and Sunayana Sitaram and Xing Xie}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sIsbOkQmBL} }

Large language models (LLMs) have been observed to exhibit bias towards certain cultures due to the predominance of training data obtained from English corpora. Considering that multilingual cultural data is often expensive to procure, existing methodologies address this challenge through prompt engineering or culture-specific pre-training. However, these strategies may neglect the knowledge deficiency of low-resource cultures and necessitate substantial computing resources. In this paper, we propose CultureLLM, a cost-effective solution to integrate cultural differences into LLMs. CultureLLM employs the World Value Survey (WVS) as seed data and generates semantically equivalent training data through the proposed semantic data augmentation. Utilizing only $50$ seed samples from WVS with augmented data, we fine-tune culture-specific LLMs as well as a unified model (CultureLLM-One) for $9$ cultures, encompassing both rich and low-resource languages. Extensive experiments conducted on $60$ culture-related datasets reveal that CultureLLM significantly surpasses various counterparts such as GPT-3.5 (by $8.1$\%) and Gemini Pro (by $9.5$\%), demonstrating performance comparable to or exceeding that of GPT-4. Our human study indicates that the generated samples maintain semantic equivalence to the original samples, offering an effective solution for LLMs augmentation. Code is released at https://github.com/Scarelette/CultureLLM.

CultureLLM: Incorporating Cultural Differences into Large Language Models

NeurIPS.cc/2024/Conference

2402.10946

poster

https://openreview.net/forum?id=sGvZyV2iqN

@inproceedings{ nikolaev2024hairfastgan, title={HairFast{GAN}: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach}, author={Maxim Nikolaev and Mikhail Kuznetsov and Dmitry Vetrov and Aibek Alanov}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sGvZyV2iqN} }

Our paper addresses the complex task of transferring a hairstyle from a reference image to an input photo for virtual hair try-on. This task is challenging due to the need to adapt to various photo poses, the sensitivity of hairstyles, and the lack of objective metrics. The current state of the art hairstyle transfer methods use an optimization process for different parts of the approach, making them inexcusably slow. At the same time, faster encoder-based models are of very low quality because they either operate in StyleGAN's W+ space or use other low-dimensional image generators. Additionally, both approaches have a problem with hairstyle transfer when the source pose is very different from the target pose, because they either don't consider the pose at all or deal with it inefficiently. In our paper, we present the HairFast model, which uniquely solves these problems and achieves high resolution, near real-time performance, and superior reconstruction compared to optimization problem-based methods. Our solution includes a new architecture operating in the FS latent space of StyleGAN, an enhanced inpainting approach, and improved encoders for better alignment, color transfer, and a new encoder for post-processing. The effectiveness of our approach is demonstrated on realism metrics after random hairstyle transfer and reconstruction when the original hairstyle is transferred. In the most difficult scenario of transferring both shape and color of a hairstyle from different images, our method performs in less than a second on the Nvidia V100.

HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach

NeurIPS.cc/2024/Conference

2404.01094

https://huggingface.co/papers/2404.01094

poster

https://openreview.net/forum?id=sFaFDcVNbW

@inproceedings{ hyun2024gsgan, title={{GSGAN}: Adversarial Learning for Hierarchical Generation of 3D Gaussian Splats}, author={Sangeek Hyun and Jae-Pil Heo}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sFaFDcVNbW} }

Most advances in 3D Generative Adversarial Networks (3D GANs) largely depend on ray casting-based volume rendering, which incurs demanding rendering costs. One promising alternative is rasterization-based 3D Gaussian Splatting (3D-GS), providing a much faster rendering speed and explicit 3D representation. In this paper, we exploit Gaussian as a 3D representation for 3D GANs by leveraging its efficient and explicit characteristics. However, in an adversarial framework, we observe that a na\"ive generator architecture suffers from training instability and lacks the capability to adjust the scale of Gaussians. This leads to model divergence and visual artifacts due to the absence of proper guidance for initialized positions of Gaussians and densification to manage their scales adaptively. To address these issues, we introduce GSGAN, a generator architecture with a hierarchical multi-scale Gaussian representation that effectively regularizes the position and scale of generated Gaussians. Specifically, we design a hierarchy of Gaussians where finer-level Gaussians are parameterized by their coarser-level counterparts; the position of finer-level Gaussians would be located near their coarser-level counterparts, and the scale would monotonically decrease as the level becomes finer, modeling both coarse and fine details of the 3D scene. Experimental results demonstrate that ours achieves a significantly faster rendering speed (×100) compared to state-of-the-art 3D consistent GANs with comparable 3D generation capability.

GSGAN: Adversarial Learning for Hierarchical Generation of 3D Gaussian Splats

NeurIPS.cc/2024/Conference

2406.02968

https://huggingface.co/papers/2406.02968

poster

https://openreview.net/forum?id=sEpSxteEKJ

@inproceedings{ brenner2024almostlinear, title={Almost-Linear {RNN}s Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction}, author={Manuel Brenner and Christoph J{\"u}rgen Hemmer and Zahra Monfared and Daniel Durstewitz}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sEpSxteEKJ} }

Dynamical systems theory (DST) is fundamental for many areas of science and engineering. It can provide deep insights into the behavior of systems evolving in time, as typically described by differential or recursive equations. A common approach to facilitate mathematical tractability and interpretability of DS models involves decomposing nonlinear DS into multiple linear DS combined by switching manifolds, i.e. piecewise linear (PWL) systems. PWL models are popular in engineering and a frequent choice in mathematics for analyzing the topological properties of DS. However, hand-crafting such models is tedious and only possible for very low-dimensional scenarios, while inferring them from data usually gives rise to unnecessarily complex representations with very many linear subregions. Here we introduce Almost-Linear Recurrent Neural Networks (AL-RNNs) which automatically and robustly produce most parsimonious PWL representations of DS from time series data, using as few PWL nonlinearities as possible. AL-RNNs can be efficiently trained with any SOTA algorithm for dynamical systems reconstruction (DSR), and naturally give rise to a symbolic encoding of the underlying DS that provably preserves important topological properties. We show that for the Lorenz and Rössler systems, AL-RNNs derive, in a purely data-driven way, the known topologically minimal PWL representations of the corresponding chaotic attractors. We further illustrate on two challenging empirical datasets that interpretable symbolic encodings of the dynamics can be achieved, tremendously facilitating mathematical and computational analysis of the underlying systems.

Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction

NeurIPS.cc/2024/Conference

2410.14240

poster

https://openreview.net/forum?id=sABwo1ZTFi

@inproceedings{ yu2024generalizablity, title={Generalizablity of Memorization Neural Network}, author={Lijia Yu and Xiao-Shan Gao and Lijun Zhang and Yibo Miao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=sABwo1ZTFi} }

The neural network memorization problem is to study the expressive power of neural networks to interpolate a finite dataset. Although memorization is widely believed to have a close relationship with the strong generalizability of deep learning when using overparameterized models, to the best of our knowledge, there exists no theoretical study on the generalizability of memorization neural networks. In this paper, we give the first theoretical analysis of this topic. Since using i.i.d. training data is a necessary condition for a learning algorithm to be generalizable, memorization and its generalization theory for i.i.d. datasets are developed under mild conditions on the data distribution. First, algorithms are given to construct memorization networks for an i.i.d. dataset, which have the smallest number of parameters and even a constant number of parameters. Second, we show that, in order for the memorization networks to be generalizable, the width of the network must be at least equal to the dimension of the data, which implies that the existing memorization networks with an optimal number of parameters are not generalizable. Third, a lower bound for the sample complexity of general memorization algorithms and the exact sample complexity for memorization algorithms with constant number of parameters are given. As a consequence, it is shown that there exist data distributions such that, to be generalizable for them, the memorization network must have an exponential number of parameters in the data dimension. Finally, an efficient and generalizable memorization algorithm is given when the number of training samples is greater than the efficient memorization sample complexity of the data distribution.

Generalizablity of Memorization Neural Network

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=s8Pxz7cvHT

@inproceedings{ li2024advad, title={Adv{AD}: Exploring Non-Parametric Diffusion for Imperceptible Adversarial Attacks}, author={Jin Li and Ziqiang He and Anwei Luo and Jian-Fang Hu and Z. Jane Wang and Xiangui Kang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s8Pxz7cvHT} }

Imperceptible adversarial attacks aim to fool DNNs by adding imperceptible perturbation to the input data. Previous methods typically improve the imperceptibility of attacks by integrating common attack paradigms with specifically designed perception-based losses or the capabilities of generative models. In this paper, we propose Adversarial Attacks in Diffusion (AdvAD), a novel modeling framework distinct from existing attack paradigms. AdvAD innovatively conceptualizes attacking as a non-parametric diffusion process by theoretically exploring basic modeling approach rather than using the denoising or generation abilities of regular diffusion models requiring neural networks. At each step, much subtler yet effective adversarial guidance is crafted using only the attacked model without any additional network, which gradually leads the end of diffusion process from the original image to a desired imperceptible adversarial example. Grounded in a solid theoretical foundation of the proposed non-parametric diffusion process, AdvAD achieves high attack efficacy and imperceptibility with intrinsically lower overall perturbation strength. Additionally, an enhanced version AdvAD-X is proposed to evaluate the extreme of our novel framework under an ideal scenario. Extensive experiments demonstrate the effectiveness of the proposed AdvAD and AdvAD-X. Compared with state-of-the-art imperceptible attacks, AdvAD achieves an average of 99.9% (+17.3%) ASR with 1.34 (-0.97) $l_2$ distance, 49.74 (+4.76) PSNR and 0.9971 (+0.0043) SSIM against four prevalent DNNs with three different architectures on the ImageNet-compatible dataset. Code is available at https://github.com/XianguiKang/AdvAD.

AdvAD: Exploring Non-Parametric Diffusion for Imperceptible Adversarial Attacks

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=s63dtq0mwA

@inproceedings{ basu2024understanding, title={Understanding Information Storage and Transfer in Multi-Modal Large Language Models}, author={Samyadeep Basu and Martin Grayson and Cecily Morrison and Besmira Nushi and Soheil Feizi and Daniela Massiceti}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s63dtq0mwA} }

Understanding the mechanisms of information storage and transfer in Transformer-based models is important for driving model understanding progress. Recent work has studied these mechanisms for Large Language Models (LLMs), revealing insights on how information is stored in a model's parameters and how information flows to and from these parameters in response to specific prompts. However, these studies have not yet been extended to Multi-modal Large Language Models (MLLMs). Given their expanding capabilities and real-world use, we start by studying one aspect of these models -- how MLLMs process information in a factual visual question answering task. We use a constraint-based formulation which views a visual question as having a set of visual or textual constraints that the model's generated answer must satisfy to be correct (e.g. What movie directed by \emph{the director in this photo} has won a \emph{Golden Globe}?). Under this setting, we contribute i) a method that extends causal information tracing from pure language to the multi-modal setting, and ii) \emph{VQA-Constraints}, a test-bed of 9.7K visual questions annotated with constraints. We use these tools to study two open-source MLLMs, LLaVa and multi-modal Phi-2. Our key findings show that these MLLMs rely on MLP and self-attention blocks in much earlier layers for information storage, compared to LLMs whose mid-layer MLPs are more important. We also show that a consistent small subset of visual tokens output by the vision encoder are responsible for transferring information from the image to these causal blocks. We validate these mechanisms by introducing MultEdit a model-editing algorithm that can correct errors and insert new long-tailed information into MLLMs by targeting these causal blocks. We will publicly release our dataset and code.

Understanding Information Storage and Transfer in Multi-Modal Large Language Models

NeurIPS.cc/2024/Conference

2406.04236

poster

https://openreview.net/forum?id=s5917zor6V

@inproceedings{ zhang2024on, title={On the Curses of Future and History in Future-dependent Value Functions for Off-policy Evaluation}, author={Yuheng Zhang and Nan Jiang}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s5917zor6V} }

We study off-policy evaluation (OPE) in partially observable environments with complex observations, with the goal of developing estimators whose guarantee avoids exponential dependence on the horizon. While such estimators exist for MDPs and POMDPs can be converted to history-based MDPs, their estimation errors depend on the state-density ratio for MDPs which becomes history ratios after conversion, an exponential object. Recently, Uehara et al. [2022a] proposed future-dependent value functions as a promising framework to address this issue, where the guarantee for memoryless policies depends on the density ratio over the latent state space. However, it also depends on the boundedness of the future-dependent value function and other related quantities, which we show could be exponential-in-length and thus erasing the advantage of the method. In this paper, we discover novel coverage assumptions tailored to the structure of POMDPs, such as outcome coverage and belief coverage, which enable polynomial bounds on the aforementioned quantities. As a side product, our analyses also lead to the discovery of new algorithms with complementary properties.

On the Curses of Future and History in Future-dependent Value Functions for Off-policy Evaluation

NeurIPS.cc/2024/Conference

2402.14703

poster

https://openreview.net/forum?id=s4Wx2qXhv9

@inproceedings{ voracek2024treatment, title={Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness}, author={Vaclav Voracek}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s4Wx2qXhv9} }

Randomized smoothing is a popular certified defense against adversarial attacks. In its essence, we need to solve a problem of statistical estimation which is usually very time-consuming since we need to perform numerous (usually $10^5$) forward passes of the classifier for every point to be certified. In this paper, we review the statistical estimation problems for randomized smoothing to find out if the computational burden is necessary. In particular, we consider the (standard) task of adversarial robustness where we need to decide if a point is robust at a certain radius or not using as few samples as possible while maintaining statistical guarantees. We present estimation procedures employing confidence sequences enjoying the same statistical guarantees as the standard methods, with the optimal sample complexities for the estimation task and empirically demonstrate their good performance. Additionally, we provide a randomized version of Clopper-Pearson confidence intervals resulting in strictly stronger certificates.

Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness

NeurIPS.cc/2024/Conference

2406.17830

poster

https://openreview.net/forum?id=s3icZC2NLq

@inproceedings{ zhao2024a, title={A Nearly Optimal and Low-Switching Algorithm for Reinforcement Learning with General Function Approximation}, author={Heyang Zhao and Jiafan He and Quanquan Gu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s3icZC2NLq} }

The exploration-exploitation dilemma has been a central challenge in reinforcement learning (RL) with complex model classes. In this paper, we propose a new algorithm, Monotonic Q-Learning with Upper Confidence Bound (MQL-UCB) for RL with general function approximation. Our key algorithmic design includes (1) a general deterministic policy-switching strategy that achieves low switching cost, (2) a monotonic value function structure with carefully controlled function class complexity, and (3) a variance-weighted regression scheme that exploits historical trajectories with high data efficiency. MQL-UCB achieves minimax optimal regret of $\tilde{O}(d\sqrt{HK})$ when $K$ is sufficiently large and near-optimal policy switching cost of $\tilde{O}(dH)$, with $d$ being the eluder dimension of the function class, $H$ being the planning horizon, and $K$ being the number of episodes. Our work sheds light on designing provably sample-efficient and deployment-efficient Q-learning with nonlinear function approximation.

A Nearly Optimal and Low-Switching Algorithm for Reinforcement Learning with General Function Approximation

NeurIPS.cc/2024/Conference

2311.15238

poster

https://openreview.net/forum?id=s2hA6Bz3LE

@inproceedings{ smerkous2024enhancing, title={Enhancing Diversity in Bayesian Deep Learning via Hyperspherical Energy Minimization of {CKA}}, author={David Smerkous and Qinxun Bai and Li Fuxin}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s2hA6Bz3LE} }

Particle-based Bayesian deep learning often requires a similarity metric to compare two networks. However, naive similarity metrics lack permutation invariance and are inappropriate for comparing networks. Centered Kernel Alignment (CKA) on feature kernels has been proposed to compare deep networks but has not been used as an optimization objective in Bayesian deep learning. In this paper, we explore the use of CKA in Bayesian deep learning to generate diverse ensembles and hypernetworks that output a network posterior. Noting that CKA projects kernels onto a unit hypersphere and that directly optimizing the CKA objective leads to diminishing gradients when two networks are very similar. We propose adopting the approach of hyperspherical energy (HE) on top of CKA kernels to address this drawback and improve training stability. Additionally, by leveraging CKA-based feature kernels, we derive feature repulsive terms applied to synthetically generated outlier examples. Experiments on both diverse ensembles and hypernetworks show that our approach significantly outperforms baselines in terms of uncertainty quantification in both synthetic and realistic outlier detection tasks.

Enhancing Diversity in Bayesian Deep Learning via Hyperspherical Energy Minimization of CKA

NeurIPS.cc/2024/Conference

2411.00259

poster

https://openreview.net/forum?id=s1MoH2pACa

@inproceedings{ sun2024ensir, title={Ens{IR}: An Ensemble Algorithm for Image Restoration via Gaussian Mixture Models}, author={Shangquan Sun and Wenqi Ren and Zikun Liu and Hyunhee Park and Rui Wang and Xiaochun Cao}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=s1MoH2pACa} }

Image restoration has experienced significant advancements due to the development of deep learning. Nevertheless, it encounters challenges related to ill-posed problems, resulting in deviations between single model predictions and ground-truths. Ensemble learning, as a powerful machine learning technique, aims to address these deviations by combining the predictions of multiple base models. Most existing works adopt ensemble learning during the design of restoration models, while only limited research focuses on the inference-stage ensemble of pre-trained restoration models. Regression-based methods fail to enable efficient inference, leading researchers in academia and industry to prefer averaging as their choice for post-training ensemble. To address this, we reformulate the ensemble problem of image restoration into Gaussian mixture models (GMMs) and employ an expectation maximization (EM)-based algorithm to estimate ensemble weights for aggregating prediction candidates. We estimate the range-wise ensemble weights on a reference set and store them in a lookup table (LUT) for efficient ensemble inference on the test set. Our algorithm is model-agnostic and training-free, allowing seamless integration and enhancement of various pre-trained image restoration models. It consistently outperforms regression-based methods and averaging ensemble approaches on 14 benchmarks across 3 image restoration tasks, including super-resolution, deblurring and deraining. The codes and all estimated weights have been released in Github.

EnsIR: An Ensemble Algorithm for Image Restoration via Gaussian Mixture Models

NeurIPS.cc/2024/Conference

2410.22959

poster

https://openreview.net/forum?id=rzvVm0LsyK

@inproceedings{ taniguchi2024adopt, title={{ADOPT}: Modified Adam Can Converge with Any \${\textbackslash}beta\_2\$ with the Optimal Rate}, author={Shohei Taniguchi and Keno Harada and Gouki Minegishi and Yuta Oshima and Seong Cheol Jeong and Go Nagahara and Tomoshi Iiyama and Masahiro Suzuki and Yusuke Iwasawa and Yutaka Matsuo}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=rzvVm0LsyK} }

Adam is one of the most popular optimization algorithms in deep learning. However, it is known that Adam does not converge in theory unless choosing a hyperparameter, i.e., $\beta_2$, in a problem-dependent manner. There have been many attempts to fix the non-convergence (e.g., AMSGrad), but they require an impractical assumption that the gradient noise is uniformly bounded. In this paper, we propose a new adaptive gradient method named ADOPT, which achieves the optimal convergence rate of $\mathcal{O} ( 1 / \sqrt{T} )$ with any choice of $\beta_2$ without depending on the bounded noise assumption. ADOPT addresses the non-convergence issue of Adam by removing the current gradient from the second moment estimate and changing the order of the momentum update and the normalization by the second moment estimate. We also conduct intensive numerical experiments, and verify that our ADOPT achieves superior results compared to Adam and its variants across a wide range of tasks, including image classification, generative modeling, natural language processing, and deep reinforcement learning. The implementation is available at https://github.com/iShohei220/adopt.

ADOPT: Modified Adam Can Converge with Any β_2 with the Optimal Rate

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=ry0RXTJwjy

@inproceedings{ kong2024learning, title={Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games}, author={Fanqi Kong and Yizhe Huang and Song-Chun Zhu and Siyuan Qi and Xue Feng}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=ry0RXTJwjy} }

Real-world multi-agent scenarios often involve mixed motives, demanding altruistic agents capable of self-protection against potential exploitation. However, existing approaches often struggle to achieve both objectives. In this paper, based on that empathic responses are modulated by learned social relationships between agents, we propose LASE (**L**earning to balance **A**ltruism and **S**elf-interest based on **E**mpathy), a distributed multi-agent reinforcement learning algorithm that fosters altruistic cooperation through gifting while avoiding exploitation by other agents in mixed-motive games. LASE allocates a portion of its rewards to co-players as gifts, with this allocation adapting dynamically based on the social relationship --- a metric evaluating the friendliness of co-players estimated by counterfactual reasoning. In particular, social relationship measures each co-player by comparing the estimated $Q$-function of current joint action to a counterfactual baseline which marginalizes the co-player's action, with its action distribution inferred by a perspective-taking module. Comprehensive experiments are performed in spatially and temporally extended mixed-motive games, demonstrating LASE's ability to promote group collaboration without compromising fairness and its capacity to adapt policies to various types of interactive co-players.

Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games

NeurIPS.cc/2024/Conference

2410.07863

poster

https://openreview.net/forum?id=rvBabL7DUu

@inproceedings{ cui2024faceqr, title={Face2{QR}: A Unified Framework for Aesthetic, Face-Preserving, and Scannable {QR} Code Generation}, author={Xuehao Cui and Guangyang Wu and Zhenghao Gan and Guangtao Zhai and Xiaohong Liu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=rvBabL7DUu} }

Existing methods to generate aesthetic QR codes, such as image and style transfer techniques, tend to compromise either the visual appeal or the scannability of QR codes when they incorporate human face identity. Addressing these imperfections, we present Face2QR—a novel pipeline specifically designed for generating personalized QR codes that harmoniously blend aesthetics, face identity, and scannability. Our pipeline introduces three innovative components. First, the ID-refined QR integration (IDQR) seamlessly intertwines the background styling with face ID, utilizing a unified SD-based framework with control networks. Second, the ID-aware QR ReShuffle (IDRS) effectively rectifies the conflicts between face IDs and QR patterns, rearranging QR modules to maintain the integrity of facial features without compromising scannability. Lastly, the ID-preserved Scannability Enhancement (IDSE) markedly boosts scanning robustness through latent code optimization, striking a delicate balance between face ID, aesthetic quality and QR functionality. In comprehensive experiments, Face2QR demonstrates remarkable performance, outperforming existing approaches, particularly in preserving facial recognition features within custom QR code designs.

Face2QR: A Unified Framework for Aesthetic, Face-Preserving, and Scannable QR Code Generation

NeurIPS.cc/2024/Conference

poster

https://openreview.net/forum?id=rtz4df9IF1

@inproceedings{ cunha2024optimal, title={Optimal Parallelization of Boosting}, author={Arthur da Cunha and Mikael M{\o}ller H{\o}gsgaard and Kasper Green Larsen}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=rtz4df9IF1} }

Recent works on the parallel complexity of Boosting have established strong lower bounds on the tradeoff between the number of training rounds $p$ and the total parallel work per round $t$. These works have also presented highly non-trivial parallel algorithms that shed light on different regions of this tradeoff. Despite these advancements, a significant gap persists between the theoretical lower bounds and the performance of these algorithms across much of the tradeoff space. In this work, we essentially close this gap by providing both improved lower bounds on the parallel complexity of weak-to-strong learners, and a parallel Boosting algorithm whose performance matches these bounds across the entire $p$ vs. $t$ compromise spectrum, up to logarithmic factors. Ultimately, this work settles the parallel complexity of Boosting algorithms that are nearly sample-optimal.

Optimal Parallelization of Boosting

NeurIPS.cc/2024/Conference

2408.16653

oral

https://openreview.net/forum?id=rpjh69DUX2

@inproceedings{ xu2024metareinforcement, title={Meta-Reinforcement Learning with Universal Policy Adaptation: Provable Near-Optimality under All-task Optimum Comparator}, author={Siyuan Xu and Minghui Zhu}, booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems}, year={2024}, url={https://openreview.net/forum?id=rpjh69DUX2} }

Meta-reinforcement learning (Meta-RL) has attracted attention due to its capability to enhance reinforcement learning (RL) algorithms, in terms of data efficiency and generalizability. In this paper, we develop a bilevel optimization framework for meta-RL (BO-MRL) to learn the meta-prior for task-specific policy adaptation, which implements multiple-step policy optimization on one-time data collection. Beyond existing meta-RL analyses, we provide upper bounds of the expected optimality gap over the task distribution. This metric measures the distance of the policy adaptation from the learned meta-prior to the task-specific optimum, and quantifies the model's generalizability to the task distribution. We empirically validate the correctness of the derived upper bounds and demonstrate the superior effectiveness of the proposed algorithm over benchmarks.

Meta-Reinforcement Learning with Universal Policy Adaptation: Provable Near-Optimality under All-task Optimum Comparator

NeurIPS.cc/2024/Conference

2410.09728