AI.info

William Yang Wang

Explore William Yang Wang on AI.info.

• LogicReward: Incentivizing LLM Reasoning via Step-Wise Logical Supervision
• Atomic Skills are the Prerequisite: When Reinforcement Learning Synthesizes Compositional Reasoning, and When It Only Amplifies
• Budget-Aware Tool Use Enables Effective Agent Scaling
• Do You Know About My Nation? Investigating Multilingual Language Models' Cultural Literacy Through Factual Knowledge
• Dynamic Evaluation for Oversensitivity in LLMs