Yue Wang

5 papers Soochow University, China

• First SFT, Second RL, Third UPT: Continual Improving Multi-Modal LLM Reasoning via Unsupervised Post-Training
• Martian World Model: Controllable Video Synthesis with Physically Accurate 3D Reconstructions
• Seeing the Wind from a Falling Leaf
• Thoughts Are All Over the Place: On the Underthinking of Long Reasoning Models
• Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training