LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models
Abstract
Visual instruction tuning has made considerable strides in enhancing the capabilities of Large Multimodal Models (LMMs). However, existing open LMMs largely focus on single-image tasks, their applications to multi-image scenarios remains less explored. Additionally, prior LMM research separately tackles different scenarios, leaving it impossible to generalize cross scenarios with new emerging capabilities. To this end, we introduce LLaVA-NeXT-Interleave, which simultaneously tackles Multi-image, Multi-frame (video), Multi-view (3D), and Multi-patch (single-image) scenarios in LMMs. To enable these capabilities, we regard the interleaved data format as a general template and compile the M4-Instruct dataset with 1,177.6k samples, spanning 4 primary domains with 14 tasks and 41 datasets. We also curate the LLaVA-Interleave Bench to comprehensively evaluate the multi-image performance of LMMs. Through extensive experiments, LLaVA-NeXT-Interleave achieves leading results in multi-image, video, and 3D benchmarks, while maintaining the performance of single-image tasks. Besides, our model also exhibits several emerging capabilities, e.g., transferring tasks across different settings and modalities. Code is available at https://github.com/LLaVA-VL/LLaVA-NeXT
Community
hi
@ZrrSkywalker
@Haozhangcx
, congrats on your work ๐ฅ
It would be great if you could link the model, dataset and demo with the paper. You can follow the guide here: https://huggingface.co/docs/hub/en/paper-pages#linking-a-paper-to-a-model-dataset-or-space.
I've linked the HF Transformers compatible models to the hub, you can find them in this collection: https://huggingface.co/collections/lmms-lab/llava-next-interleave-66763c55c411b340b35873d1
Models citing this paper 6
Browse 6 models citing this paperDatasets citing this paper 0
No dataset linking this paper