Lijuan Wang

4 papers

• Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuning
• SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
• VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents
• ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs