Vision-LLM-Alignment aims to implement alignment training for visual large language models (LLMs), encompassing SFT training, reward model training, and PPO/DPO training. For the integration of additional alignment algorithms or to report any arising bugs, please submit an issue.
-
[2024/10/03] We support tuning for multi-image instructions on the LLaMA-3.2-Vision. See data examples for usage.
-
[2024/09/28] 💡We support for training the LLaMA-3.2-Vision. You just need to set the
model_architecture
andtemplate
parameters to "llama-3.2-vision", and specify the LLaMA-Vision model path withfrom_checkpoint
. -
[2024/08/21] 💪We released RoVRM:A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data, which is trained and applied for human-alignment training based on this repository.[Paper][Checkpoints]
-
[2024/08/19] We support for training the LLaVA-NeXT (as known as LLaVA-1.6). You just need to set the
model_architecture
parameter to "llava_next", and specify the LLaVA-NeXT model path withfrom_checkpoint
.
Full Changelog
-
[2024/07/18] We provide a large-scale vision feedback dataset. It is a combination of the following high-quality vision feedback datasets. The dataset can be found in wangclnlp/vision-feedback-mix-binarized and wangclnlp/vision-feedback-mix-binarized-cleaned.
-
[2024/07/10] We support the direct loading of a LLaVA model in all training stages, including SFT training, RM training, and PPO/DPO training.
-
[2024/07/07] We support the direct loading of a LLaVA model during the SFT training phase. You just need to set the
model_architecture
parameter to "llava" and specify the LLaVA model path withfrom_checkpoint
. Support for this functionality during the DPO, RM training, and PPO junction phases will be introduced soon.
During the development of this system, we conducted a series of benchmark tests to evaluate and validate the system's performance. Specifically, we selected RLAIF-V as the preference dataset and LLaVA-Instruct-150K as the input instruction for the RLHF training session. In the model evaluation phase, we utilized several standard benchmarks, including MMHalBench, Object HalBench, AMBER, LLaVA-Benchmark, and MMinstruct, to conduct a more comprehensive assessment of the differences in trustworthiness and helpfulness of the vision-based LLM before and after alignment.
For training the reward model, we used the LLaVA-1.5-7B model. We performed Best-of-n sampling and RLHF (Reinforcement Learning from Human Feedback) alignment training on two models: LLaVA-1.5-7B and LLaVA-1.5-13B, respectively. The benchmarking results of the system are detailed in the figure below.
Full Results
In addition, we conducted DPO training for this system, specifically targeting the LLaVA-1.5-7B and LLaVA-1.5-13B models. The results are detailed in the following figure.
You can use anaconda/miniconda to install packages needed for this project.
pip install -r requirements.txt
Vision-LLM requires both a vision encoder and a language model. Its architecture is depicted in the figure. You can also directly employ a vision LLM after SFT, such as LLaVA-1.5/-NeXT and LLaMA-3.2-Vision-Instruction, as the actor model.
We have tentatively implemented all alignment training based on this LLaVA dataset format. Some samples can be found in the data folder.
bash run_sft.sh
bash run_rm_training.sh
bash run_dpo_training.sh
bash run_ppo_training.sh
bash run_predict.sh
LLM | Model size |
---|---|
LLaMA-2 | 7B/13B/70B |
LLaMA-3 | 8B/70B |
Vision Projector |
---|
clip-vit-large-patch14 |
clip-vit-large-patch14-336 |
LLM | Model size |
---|---|
LLaVA | 7B/13B |
LLaMA-1.5 | 7B/13B |
LLaMA-NeXT/-1.6-vicuna | 7B/13B |
LLaMA-NeXT/-1.6-mistral | 7B/13B |
Llama-3.2-Vision | 11B/90B |
Note: Other LLMs with similar architectures are also supported.
Additionally, custom model architectures can be incorporated by modifying training/utils/model/build_model.py
(loading model) and training/utils/data/DST.py
(template).
We commence by utilizing the exceptional codebase provided by DeepSpeed-VisualChat 🌹🌹🌹.
We would like to thank Yifu Huo and Yang Gan for their contributions to this work.
We thank the following papers:
[1] Ouyang, Long, et al. "Training language models to follow instructions with human feedback." Advances in neural information processing systems 35 (2022): 27730-27744.
[2] Rafailov, Rafael, et al. "Direct preference optimization: Your language model is secretly a reward model." Advances in Neural Information Processing Systems 36 (2024).
[3] Liu, Haotian, et al. "Visual instruction tuning." Advances in neural information processing systems 36 (2024).
Please cite our paper if you find the repo helpful in your work:
@misc{wang2024rovrmrobustvisualreward,
title={RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data},
author={Chenglong Wang and Yang Gan and Yifu Huo and Yongyu Mu and Murun Yang and Qiaozhi He and Tong Xiao and Chunliang Zhang and Tongran Liu and Quan Du and Di Yang and Jingbo Zhu},
year={2024},
eprint={2408.12109},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2408.12109},
}