Intertemporal Preference Steering in Qwen3 via Contrastive Activation Addition 文章

ArXiv CS.AI2026-08-05PAPERen作者: Michal Mr\'az, Justin Shenk
查看原文 →
关系图谱

Intertemporal Preference Steering in Qwen3 via Contrastive Activation Addition · 相关技术

相关技术

contrastive activation-addition steeringResidual Streamcontrastive linear probes