Jieping Ye

6 papers University of Michigan

• Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning
• Controlling Thinking Speed in Reasoning Models
• EchoShot: Multi-Shot Portrait Video Generation
• Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering
• SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models
• TokenSqueeze: Performance-Preserving Compression for Reasoning LLMs