English(EN) What Should SFT Learn? Conditional Information Gain for Generalizable Post-Training

新的CIG-SFT方法增强了LLM在后训练中的泛化能力

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-12 04:00

研究人员推出了一种新颖的大型语言模型监督微调（SFT）框架CIG-SFT。该方法采用信息论方法来估计每个示范token所携带的条件信息，量化任务本身在多大程度上降低了给定决策的不确定性。通过将此信息转换为当前策略的KL正则化投影，CIG-SFT推导出一个加权的SFT目标，在无需额外数据或在线回滚的情况下，提高了数学推理和代码生成任务的泛化能力。 AI

影响该方法可以提高LLM在推理和代码生成等复杂任务上的泛化能力。

排序理由该集群包含一篇学术论文，详细介绍了一种用于微调大型语言模型的新方法。[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 →

本文如何被排名

Signal score: 11 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了一种用于微调大型语言模型的新方法。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yuxuan Gu, Kequan Niu, Xiaocheng Feng, Yun Li, Yiqiao Yang, Bing Qin · 2026-10-12 04:00

SFT应学习什么？用于可泛化训练后调整的条件信息增益

arXiv:2610.12995v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) remains a widely used approach for post-training large language models, because fixed demonstrations can directly introduce desired reasoning strategies and behavioral patterns. Yet forcing a model to …

本文如何被排名

报道来源 [1]

SFT应学习什么？用于可泛化训练后调整的条件信息增益

相关实体

相关话题