RL4VLM
RL4VLM is an open-source project for reinforcing learning to fine-tune large visual-language models, enhancing decision-making capabilities.
What is RL4VLM?
RL4VLM is an open-source project that uses reinforcement learning to fine-tune large visual-language models, enabling them to make decisions. It builds on the LLaVA model and uses the PPO algorithm for training. The project offers detailed code libraries, setup guides, and licensing information.