Letting Tutor Personas Speak Up for LLMs: Learning Steering Vectors from Dialogue via Preference Optimization 文章

ArXiv CS.CL2026-06-03NEWSen作者: Jaewook Lee, Alexander Scarlatos, Simon Woodhead, Andrew Lan

Letting Tutor Personas Speak Up for LLMs: Learning Steering Vectors from Dialogue via Preference Optimization · 相关人物

暂无数据