The Labyrinth and the Thread: Rethinking Regularizations in Sequential Knowledge Editing for Large Language Models 文章

ArXiv CS.CL2026-05-27NEWSen作者: Zheng Wang, Kaixuan Zhang, Wanfang Chen, Jingwen Zhang, Xiaonan Lu
大语言模型
查看原文 →
关系图谱

The Labyrinth and the Thread: Rethinking Regularizations in Sequential Knowledge Editing for Large Language Models · 相关技术

相关技术

ORMODELLMlanguage modelUCTSynthetic Aperture RadarStraight-Through EstimatorSpatial Pivot-Aligned Coordinate-free Embedding (SPACE)SPAOWSOWLNATMEClarge language modelsHISForEffort Metric AttentionDiTARGANNTableCLEAR