PulseAugur
实时 06:51:35
English(EN) What Matters in On-Policy Distillation? A Perspective on Data Efficiency and Data Selection

策略内蒸馏:困难示例可提升大型语言模型的推理能力

一篇新论文探讨了在策略内蒸馏(On-Policy Distillation, OPD)对于增强大型语言模型的有效性,特别关注数据效率和数据选择。研究发现,即使是单个示例(1-shot OPD)也可能有效,而更困难的示例通常能带来更优越的性能提升。研究表明,这种改进源于复杂问题中更长的思维链(Chain-of-Thought, CoT)路径,这有助于保持与教师模型的对齐并教授批判性思维模式。基于这些发现,提出了一种仅使用困难示例的数据选择方法,仅用8个精选的困难示例就达到了与更大数据集相当的性能。 AI

影响 提出了一种更具数据效率的大型语言模型训练方法,可能降低计算成本并提高推理能力。

排序理由 学术论文,详细介绍了一种新的大型语言模型训练方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

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

策略内蒸馏:困难示例可提升大型语言模型的推理能力

本文如何被排名

Signal score
27 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
学术论文,详细介绍了一种新的大型语言模型训练方法。[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]

  1. arXiv cs.AI TIER_1 English(EN) · Zhinan Hou, Jiaqi Zhang, Xunliang Cai, Keyou You ·

    策略内蒸馏的关键是什么?数据效率与数据选择的视角

    arXiv:2609.05198v1 Announce Type: new Abstract: On-Policy Distillation (OPD) has emerged as a widely adopted post-training paradigm for enhancing large language models in reasoning domains. However, the data-centric mechanisms in OPD remain relatively underexplored. This paper pr…