跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1295新增论文
+1026其中 AI 相关
+111已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-17 04:58 · 全站最近入库:2026-11-17 16:10各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 16:22 — 2026-11-17 16:22(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-17 04:58最近发现新增版本:2026-11-17 16:10库内论文最新更新时间:2026-11-17 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

25,122 已收录论文20,967 AI 相关7,309 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 1 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-17 04:57(北京时间)
新论文 · v1arXiv:2610.35586

Linear Probes Transfer from Base to Distilled Language Models despite Rotation

Marios Tsatsos

原文摘要

Linear probes, classifiers that read a language model's internal activations, are widely used to detect harmful requests. They are cheap to train and run, and an auditor who has trained one would like to reuse it when the model is replaced by a successor rather than retrain it for every release. We test how well probes can be reused. For four base models (Qwen2.5-Math-7B, Llama-3.1-8B, Qwen2.5-14B, Qwen2.5-32B) we train a probe on the base model with the XSTest prompt set and apply it, unchanged, to two successors of that base: the DeepSeek-R1 distilled reasoning model and the developer's instruction-tuned chat model. We find that the successors rotate the direction separating unsafe from safe prompts by 17$^\circ$ to 35$^\circ$, yet on XSTest the transferred probe's AUC changes by only $-0.011$ to $+0.027$, and on an independent OR-Bench evaluation, never used for training, by $-0.015$ to $+0.053$. On XSTest, whose safe prompts deliberately contain alarming words, the probes beat a classifier that sees only the prompt text by 0.15 to 0.20 AUC. On the OR-Bench prompts that margin shrinks to 0.03 to 0.20, and the probes flag 29\% to 42\% of prompts that sound harmful but are benign, so part of what they read is the appearance of harm rather than harm itself. A probe trained on a base model can thus be reused, in these families, after distillation and after instruction tuning; how much it reads beyond textual cues depends on the test set.

cs.AI · cs.CL · cs.LG