English(EN) MicroMouse: Masking Harmful Supervision in Long CoT via Relative Potential

新的MicroMouse框架掩盖了大型语言模型推理中的有害监督

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-04 05:00

研究人员开发了一个名为MicroMouse的新框架，通过识别和掩盖长链式思考（CoT）推理过程中的“有害监督”来改进大型语言模型（LLMs）的训练。MicroMouse受解迷宫的启发，使用“相对势能”（REP）指标来衡量每个推理步骤与正确答案的接近程度，并将其与一个看似正确的错误答案进行对比。这种方法有助于在监督微调（SFT）过程中过滤掉冗余或错误的步骤，从而提高LLM推理的准确性和效率。 AI

影响这项研究通过提高复杂推理任务训练数据的质量，有望带来更准确、更高效的大型语言模型。

排序理由该集群包含一篇详细介绍改进大型语言模型训练新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 43 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍改进大型语言模型训练新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Xiaowei Zhu, Hengzhu Tang, Zhenyu Zhang, Shuohuan Wang, Yubing Ren, Yanan Cao · 2026-11-04 05:00

MicroMouse：在长CoT中通过相对势能掩盖有害监督

arXiv:2611.02710v1 Announce Type: new Abstract: Even successful long CoTs may contain harmful supervision, motivating the need for methods that can automatically identify and mask such signals during SFT. Inspired by the MicroMouse maze-solving process, we view a long CoT as a se…

本文如何被排名

报道来源 [1]

MicroMouse：在长CoT中通过相对势能掩盖有害监督

相关实体

相关话题