This repository hosts the code, data, and model weight of RLHF-V, a novel framework that aligns Multimodal Large Language Models (MLLMs) behavior through fine-grained correctional human feedback. We ...

Do not do unto others what you would not have others do unto you. This is a line in The Analects of Confucius. It advises one to show empathy toward others, put oneself in another's shoes, and be ...