From Expressivity to Sample Complexity: Narrow Teachers for Transformers via C-RASP 文章

ArXiv CS.CL2026-07-14PAPERen作者: Michael Rizvi-Martel, Satwik Bhattamishra, Guillaume Rabusseau, Michael Hahn
大语言模型
查看原文 →
关系图谱

From Expressivity to Sample Complexity: Narrow Teachers for Transformers via C-RASP · 相关技术

相关技术

Transformer modelattention-based modelsC-RASPLarge language modeltransformers