metadata

license: apache-2.0
datasets:
  - Yirany/UniMM-Chat
  - HaoyeZhang/RLHF-V-Dataset
language:
  - en
library_name: transformers

Model Card for RLHF-V

Project Page|GitHub |Demo|Paper

RLHF-V is an open-source multimodal large language model with the lowest hallucination rate on both long-form instructions and short-form questions.

RLHF-V is trained on RLHF-V-Dataset, which contains fine-grained segment-level human corrections on diverse instructions. The base model is trained on UniMM-Chat, which is a high-quality knowledge-intensive SFT dataset. We introduce a new method Dense Direct Preference Optimization (DDPO) that can make better use of the fine-grained annotations.

For more details, please refer to our paper.

Model Details

Model Description

Trained from model: Based on Vicuna-13B
Trained on data: RLHF-V-Dataset

Model Sources

Project Page: https://rlhf-v.github.io
GitHub Repository: https://github.com/RLHF-V/RLHF-V
Demo: http://120.92.209.146:8081
Paper: https://arxiv.org/abs/2312.00849

Performance

Low hallucination rate while being informative:

More resistant to over-generalization, even compared to GPT-4V:

Citation

If you find RLHF-V is useful in your work, please cite it with:

@article{2023rlhf-v,
  author      = {Tianyu Yu and Yuan Yao and Haoye Zhang and Taiwen He and Yifeng Han and Ganqu Cui and Jinyi Hu and Zhiyuan Liu and Hai-Tao Zheng and Maosong Sun and Tat-Seng Chua},
  title       = {RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback},
  journal      = {arxiv},
  year         = {2023},
}