This paper uncovers a stable relationship between internal response representations and problem-level empirical accuracy and proposes SelectTTRL, a representation-guided data selection method that ranks test problems using COS and applies TTRL only to a selected percentile interval.
Experiments show that Learner2Skill more faithfully reproduces fine-grained learner behavior while reducing overall token cost, and that the same constructed Skills can be effectively reused across different LLM executors.
DuoOPD is introduced, in which the student's outcome sets the direction of feedback and the joint teacher-student outcome decides how the teacher supports it: when only the teacher succeeds, its verified answer becomes context for scoring the student's failed response.
GRACE is proposed, a Graph-based Rationalized CoT Compression framework that transforms long CoTs into reasoning graphs to enable efficient CoT compression, where reasoning steps are represented as nodes and semantic dependencies as edges.
Yizhi Wang, Li-Nan Yue, Deng-Bao Wang et al.· Proceedings of the 32nd ACM...· 0 citations
LightTIR, a dual-penalty reward framework, is proposed to achieve efficient TIR and can reduce redundancy and trajectory expansion while maintaining answer correctness, achieving more efficient RL-based TIR.
Yichen Xiao, Siyu Gong, Li-Nan Yue· Proceedings of the 32nd ACM...· 0 citations
Recent methods using Reinforcement Learning (RL) have improved Tool-Integrated Reasoning (TIR) by training large language models to learn end-to-end policies for multi-step tool usage, enabling them to solve complex tasks more effectively. Despite these advances, existing methods often suffer from overthinking at both...
Yichen Xiao, Siyu Gong, Linan Yue· Proceedings of the 32nd ACM...· 0 citations
Recent advances in multimodal large reasoning models (MLRMs) have demonstrated impressive capabilities on complex multimodal tasks, yet their reliance on long Chain-of-Thoughts (CoTs) often leads to redundant reasoning and high computational cost. Existing chain-based distillation and refinement approaches alleviate re...
Yizhi Wang, Linan Yue, Deng-Bao Wang et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.