推理大语言模型的监督微调（SFT）综述
Supervised Fine-Tuning for Reasoning Large Language Models: A Survey
AI 导读
一篇被 EMNLP 2026 接收的 arXiv 综述系统梳理了面向推理的大语言模型监督微调（SFT）研究。综述从两条主线展开：方法上按数据中心与算法创新分类回顾 SFT 流程改进；分析研究则沿数据特性、优化动态与机理洞察三个维度重组，并总结当前研究焦点、瓶颈与未来方向。
正文 · 原文
Abstract:Large reasoning models have recently demonstrated remarkable capabilities in solving complex tasks, where supervised fine-tuning (SFT) on long chain-of-thought data serves as a crucial foundation for eliciting and enhancing their reasoning abilities. Despite rapid progress in both improving and analyzing reasoning-oriented SFT, the field still lacks a systematic survey that consolidates its fast-growing literature. To fill this gap, we present a comprehensive review of recent advancements in reasoning SFT, examining the literature through the dual lenses of methodological design and analytical investigation. First, we review methodological improvements across the SFT pipeline and categorize them into data-centric approaches and algorithm-centric innovations. Second, we reorganize analytical studies along three dimensions: data characteristics, optimization dynamics, and mechanistic insights. Finally, by synthesizing current research focuses and remaining bottlenecks, we outline promising future directions for reasoning SFT. We hope this survey deepens the understanding of reasoning SFT and paves the way for advanced reasoning models.
| Comments: | Accepted to EMNLP 2026 Conference |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.37681 [cs.CL] |
| (or arXiv:2610.37681v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.37681 arXiv-issued DOI via DataCite |
Submission history
From: Yongcan Yu [view email]
[v1]
Tue, 13 Oct 2026 03:22:55 UTC (81 KB)
来源：arXiv cs.AI · arxiv.org