Picture for Yifeng Han

Yifeng Han

RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback

Add code
Dec 01, 2023
Viaarxiv icon