LongLLaVA-53B-A13B / README.md
nielsr's picture
nielsr HF staff
Add metadata tags
9fbe531 verified
|
raw
history blame
1.57 kB
metadata
license: mit
library_name: transformers
pipeline_tag: image-text-to-text

header

๐Ÿ“ƒ Paper โ€ข ๐ŸŒ Demo โ€ข ๐Ÿ“ƒ LongLLaVA

efficiency

๐ŸŒˆ Update

  • [2024.09.05] LongLLaVA repo is published๏ผ๐ŸŽ‰

Architecture

Click to view the architecture image

Architecture Image

Results

Click to view the Results
  • Main Results Main Results
  • Diagnostic Results Diagnostic Results
  • Video-NIAH Video-NIAH

Results reproduction

Data DownLoad and Construction

Dataset Taxonomy

Dataset

Dataset DownLoading and Construction

Coming Soon~

Citation

@misc{wang2024longllavascalingmultimodalllms,
      title={LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via Hybrid Architecture}, 
      author={Xidong Wang and Dingjie Song and Shunian Chen and Chen Zhang and Benyou Wang},
      year={2024},
      eprint={2409.02889},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2409.02889}, 
}