4 papers
Verify, Repair, Repeat, or Stop? Robust Stopping for Noisy Verify-Repair Loops in LLM Agents
Yitao Wu, Si Shen, Rui Yang +2
Verify-repair loops are a standard means for large language model (LLM) agents to correct faulty plans in code generation, mathematical reasoning, and tool use. When both the verif…
LLM-Metrics: Measuring Research Impact Through Large Language Model Memory
Si Shen, Wenhua Zhao, Danhao Zhu
Citation counts remain the dominant metric for assessing research impact, yet they suffer from well-documented limitations: temporal lag, disciplinary bias, and Matthew effects. He…
Mitigating Think-Answer Mismatch in LLM Reasoning Through Noise-Aware Advantage Reweighting
Si Shen, Peijun Shen, Wenhua Zhao +1
Group-Relative Policy Optimization (GRPO) is a key technique for training large reasoning models, yet it suffers from a critical vulnerability: the \emph{Think-Answer Mismatch}, wh…
Long Is More Important Than Difficult for Training Reasoning Models
Si Shen, Fei Huang, Zhixiao Zhao +3
Difficult problems, which often result in long reasoning traces, are widely recognized as key factors for enhancing the performance of reasoning models. However, such high-challeng…