Michael Qizhe Shieh

MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use

GEM: A Gym for Generalist LLMs

ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning

LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewards

S-ATM: Self-Boosting Visual Reasoning via Adaptive Token Merging

Gradually Compacting Large Language Models for Reasoning Like a Boiling Frog

SynthRL: Scaling Visual Reasoning with Verifiable Data Synthesis

NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation

RAPID: Long-Context Inference with Retrieval-Augmented Speculative Decoding

Efficient Process Reward Model Training via Active Learning

The Emergence of Abstract Thought in Large Language Models Beyond Any Language

Unnatural Languages Are Not Bugs but Features for LLMs