While Large Reasoning Models (LRMs) have demonstrated remarkable capabilities through explicit Chain-of-Thought (CoT) generation, they frequently suffer from “overthinking”. In this work, we bridge this gap by introducing Token-level Marginal Utility, which quantifies the per-token log-probability gain of the ground-truth answer. Leveraging this dense supervision signal, we propose MUTO (Marginal Utility Guided Thinking Optimization), a unified training framework designed to synthesize concise reasoning chains. Rather than relying only on coarse trajectory-level length control, MUTO identifies tokens that reduce the model’s likelihood of the correct answer and penalizes such negative-utility reasoning, yielding concise yet effective CoT trajectories. Experiments on DeepSeek-R1-Distill-Qwen backbones (1.5B and 7B) across six math reasoning benchmarks show that MUTO yields a markedly better efficiency-accuracy Pareto frontier. It reduces average token usage by 87.1% at 1.5B while improving accuracy by 2.3%, and cuts tokens by 80.2% at 7B with only -0.1% accuracy change, achieving the best length-normalized accuracy among baselines.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Bin Xu, Yu Bai Beijing Academy of Artificial Intelligence, Huashan Sun, Yiguan Lin Beijing Information Science and Technology University, Siming Liu, Xinyue Liang, Yaolin Li Beijing Institute of Technology, Zhuangzhi Dong 等 13 位作者
As large language models continue to advance, their application in educational contexts remains underexplored and under-optimized. In this paper, we address this gap by introducing the first diverse benchmark tailored for educational scenarios, incorporating synthetic data containing 9 major scenarios and over 4,000 distinct educational contexts. To enable comprehensive assessment, we propose a set of multi-dimensional evaluation metrics that cover 12 critical aspects relevant to both teachers and students. We further apply human annotation to ensure the effectiveness of the model-generated evaluation responses. Additionally, we succeed to train a relatively small-scale model on our constructed dataset and demonstrate that it can achieve performance comparable to state-of-the-art large models (e.g., Deepseek V3, Qwen Max) on the test set. Overall, this work provides a practical foundation for the development and evaluation of education-oriented language models.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表