3 papers
cs.LG2026
CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs
Yongcheng Zeng, Zexu Sun, Bokai Ji +7
Curriculum learning plays a crucial role in enhancing the training efficiency of large language models (LLMs) on reasoning tasks. However, existing methods often fail to adequately…
cs.CL2026
AgentSkiller: Scaling Generalist Agent Intelligence through Semantically Integrated Cross-Domain Data Synthesis
Zexu Sun, Bokai Ji, Hengyi Cai +4
Large Language Model agents demonstrate potential in solving real-world problems via tools, yet generalist intelligence is bottlenecked by scarce high-quality, long-horizon data. E…
cs.LG2025
Cog-Rethinker: Hierarchical Metacognitive Reinforcement Learning for LLM Reasoning
Zexu Sun, Yongcheng Zeng, Erxue Min +3
Contemporary progress in large language models (LLMs) has revealed notable inferential capacities via reinforcement learning (RL) employing verifiable reward, facilitating the deve…