Hong Cheng
2 papers · Latest:
Machine Learning
Dynamic Skill Lifecycle Management for Agentic Reinforcement Learning
SLIM dynamically manages external skills for LLM agents in RL, optimizing their active skill set for improved task performance.
2605.10923
Machine LearningUniPool: A Globally Shared Expert Pool for Mixture-of-Experts
UniPool introduces a globally shared expert pool for MoE architectures, significantly improving efficiency and performance by decoupling expert capacity from model depth.
2605.06665
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.