作者位次 #2/7ACL 2026Findings· Findings of ACL
Beyond Rejection Sampling: Trajectory Fusion for Scaling Mathematical Reasoning
Jie Deng · Hanshuang Tong · Jun Li · Shining Liang · Ning Wu · Hongzhi Li · Yutao Xie
摘要
Large language models (LLMs) have made impressive strides in mathematical reasoning, often fine-tuned using rejection sampling, which retains only correct reasoning trajectories. While effective, this paradigm treats supervision as a binary filter that systematically excludes teacher-generated er...