4 papers
Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoning
Xinhao Yao, Ruifeng Ren, Yun Liao +2
Chain-of-Thought (CoT) training has markedly advanced the reasoning capabilities of large language models (LLMs), yet the mechanisms by which CoT training enhances generalization r…
Learn More with Less: Uncertainty Consistency Guided Query Selection for RLVR
Hao Yi, Yulan Hu, Xin Li +3
Large Language Models (LLMs) have recently improved mathematical reasoning through Reinforcement Learning with Verifiable Reward (RLVR). However, existing RLVR algorithms require l…
Gradient Coupling: The Hidden Barrier to Generalization in Agentic Reinforcement Learning
Jingyu Liu, Xiaopeng Wu, Jingquan Peng +4
Reinforcement learning (RL) is a dominant paradigm for training autonomous agents, yet these agents often exhibit poor generalization, failing to adapt to scenarios not seen during…
Put the Space of LoRA Initialization to the Extreme to Preserve Pre-trained Knowledge
Pengwei Tang, Xiaolin Hu, Yong Liu +4
Low-Rank Adaptation (LoRA) is the leading parameter-efficient fine-tuning method for Large Language Models (LLMs), but it still suffers from catastrophic forgetting. Recent work ha…