Tool Calling is Linearly Readable and Steerable in Language Models 文章
ArXiv CS.CL2026-05-27NEWSen作者: Zekun Wu (University College London), Ze Wang (University College London), Seonglae Cho (Holistic AI), Yufei Yang (Imperial College London), Adriano Koshiyama (University College London), Sahan Bulathwela (University College London), Maria Perez-Ortiz (University College London)
Tool Calling is Linearly Readable and Steerable in Language Models · 相关技术
相关技术
LLaMAICEAPIAgentANNARGDAPTEffort Metric AttentionFOLForGEMMHISJSONMulti-Modal Interactive Agent LayerODEOWSPLAReferring expression comprehension (REC)SPASSESpatial Pivot-Aligned Coordinate-free Embedding (SPACE)Straight-Through EstimatorTEETool CallingToolsUCTdivide-and-conquer partitioningexeclanguage modeltool-calling研究法辛烷值