search

swordlidev / Efficient-Multimodal-LLMs-Survey

Efficient Multimodal Large Language Models: A Survey
Apache License 2.0
281 stars 12 forks source link

Wonderful Survey! And about Related Work #1

Closed sdc17 closed 6 months ago

sdc17 commented 6 months ago

Hi, thank you for sharing this wonderful survey! It is very detailed and enlightening. Do you have any plans about incorporating the following works:

  1. CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers. arXiv 2023. To be appeared in ICML 2024.
  2. UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers. ICML 2023.

, which are related to the topic of efficient VLMs.

sdc17 commented 6 months ago

By the way, it seems the paper link in README is incorrect:

https://github.com/lijiannuist/Efficient-Multimodal-LLMs-Survey/blob/39a58098ddac53ecacc4fc53e8b549fc1a8ffa3b/README.md?plain=1#L3

, the arXiv url given here is incomplete.

swordlidev commented 6 months ago

Great works, We will include these papers in next version.

swordlidev commented 6 months ago

By the way, it seems the paper link in README is incorrect:

https://github.com/lijiannuist/Efficient-Multimodal-LLMs-Survey/blob/39a58098ddac53ecacc4fc53e8b549fc1a8ffa3b/README.md?plain=1#L3

, the arXiv url given here is incomplete.

Thank you for your reminder