RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval Paper • 2409.10516 • Published Sep 16 • 37
Source2Synth: Synthetic Data Generation and Curation Grounded in Real Data Sources Paper • 2409.08239 • Published Sep 12 • 16
MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark Paper • 2409.02813 • Published Sep 4 • 28
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA Paper • 2409.02897 • Published Sep 4 • 44
SWE-bench-java: A GitHub Issue Resolving Benchmark for Java Paper • 2408.14354 • Published Aug 26 • 40
CLIPA-v2: Scaling CLIP Training with 81.1% Zero-shot ImageNet Accuracy within a \10,000 Budget; An Extra 4,000 Unlocks 81.8% Accuracy Paper • 2306.15658 • Published Jun 27, 2023 • 12