Learning to Adapt SFT Data for Better Reasoning Generalization 文章

ArXiv CS.CL2026-05-27NEWSen作者: Lisong Sun, Li Wang, Chen Zhang, Jinyang Wu, Kui Zhang, Tianhao Peng, Wenjun Wu
大语言模型
查看原文 →
关系图谱

Learning to Adapt SFT Data for Better Reasoning Generalization · 相关技术

相关技术

ginORMTTEODEPLALLMlanguage model远程代码执行（RCE）后训练target modelreinforcement learningSupervised Fine-TuningSFTReferring expression comprehension (REC)Parts-of-Speech (POS) tagsNarrative Abstraction BenchmarkLMMlarge language modelsHISForFFIEffort Metric AttentionENADenseDARTDAPTARGANN