Skip to content
View AdaptVision's full-sized avatar
  • Joined Nov 15, 2025

Block or report AdaptVision

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AdaptVision/README.md

AdaptVision: Efficient Vision-Language Models via Adaptive Visual Acquisition

Release

Contents

Installation

The environment follows the Verl.

git clone https://github.com/AdaptVision/AdaptVision.git
conda create -n adaptvision python=3.11 -y
conda activate adaptvision
# veRL
pip3 install -e . 
# flash-attn
pip3 install flash-attn==2.7.3 --no-build-isolation

pip install transformers==4.51.0
pip install math_verify
pip install ray[default]
pip install tensordict==0.6.2
pip install qwen_vl_utils

Train

Data Preparation

# train file
huggingface-cli download --repo-type dataset --resume-download Senqiao/VisionThink-Smart-Train --local-dir datasets/VisionThink-Smart-Train

# val file
huggingface-cli download --repo-type dataset --resume-download Senqiao/VisionThink-Smart-Val --local-dir datasets/VisionThink-Smart-Val

Train AdaptVision via Reinforcement Learning

To use GPT as the reward model, first set the following environment variables:

AZURE_API_KEY=
AZURE_ENDPOINT=
AZURE_API_VERSION=

Run AdaptVision Training:

bash scripts/run_adaptvision.sh

Evaluation

We use lmms-eval to evaluate our model. Setup the evaulation environment by following instructions here.

We provide the evaluation code detail in scripts/vllm_adaptvision.py.

Citation

If you find this project useful in your research, please consider citing:

@article{lin2025adaptvision,
  title={AdaptVision: Efficient Vision-Language Models via Adaptive Visual Acquisition},
  author={Lin, Zichuan and Liu, Yicheng and Yang, Yang and Tao, Lvfang and Ye, Deheng},
  journal={arXiv preprint arXiv:2512.03794},
  year={2025}
}

Acknowledgement

We would like to thank the following repos for their great work:

License

  • AdaptVision is licensed under the Apache License 2.0.

Popular repositories Loading

  1. AdaptVision AdaptVision Public

    Python 4 1

  2. AdaptVision.github.io AdaptVision.github.io Public

    HTML