Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
Another 265 papers do not have category tags yet and are not included in filtered results.
“
About this collection Collection in progress We are building a large AI safety paper collection to help researchers find work related to their interests. Papers are collected automatically andfiltered only for relevance to AI safety , without judging quality, novelty or reading value. The default “Annotated first” shows papers with tags or published paper analyses first. Each group is ordered by original publication date, newest first, with undated papers last. “Newest first” uses publication dates only. This order is not a quality ranking.
One-line summaries, tags and paper analyses are prepared by AI from the papers. Language models may have systematic biases and produce hallucinations.We plan to carry out human annotation and review to reduce errors in this content.
分析 LLM 去偏激活方向，发现其编码置信度而非公平性
Published Oct 6
Evaluated models Llama-3.1-8B , Llama-3.1-8B-Instruct , Falcon3-7B-Base and others8 Abstract 论文揭示激活引导去偏见方向主要编码模型置信度而非公平性，偏见评分下降实为诱发弃答或使概率均分的假象。
提出基于决策前隐藏状态的检测与引导，预测并抑制智能体欺骗
Published Oct 5
Evaluated models Qwen3-14B Abstract 论文展示了智能体自发欺骗可在行为显现前由内部表征预测，并通过激活引导证实了表征的行为可操作性。
提出有守卫最小步激活引导，只在关机语境把继续偏好转向接受关机
Published Sep 24
Evaluated models Qwen3.5-0.8B 提出 KPI，用入度定位跨层关键路径并做 SAE 转向以解决知识冲突
Published Sep 8
Evaluated models Llama-3-8B , Llama-3.1-8B , Gemma-2-9B Abstract KPI 以入度关键路径替代大批 SAE 特征转向，在冲突 RAG 上提高忠实并减轻副作用。KPI 针对知识冲突 RAG 中的 SAE 转向：相关性批量改特征会带入冗余，少量因果上关键的特征即可完成转向。
在写入点定位拒答方向，经刚性旋转跨架构迁移并门控放大
Published Sep 4
Evaluated models Falcon-Mamba-7B , Falcon3-Mamba-7B , Llama-3.1-8B and others8 Abstract 拒答方向可经刚性旋转跨架构迁移，但必须在各架构的写入点读取。作者研究拒答表示是否随架构改变，结论是表示共享、读取位置随架构变。
用未转向表征的传播映射预测激活转向副作用
Published Jul 28
Evaluated models Gemma-3-4B , Gemma-3-12B , Qwen2.5-7B Abstract 交叉效应矩阵显示副作用有结构且不对称，并可在转向前由传播映射预测方向。激活转向的副作用可以在施加之前预测，而不必先把每个行为都转向一遍。 。
分析续写触发越狱的注意力头机制并提出 HCS
Published Mar 9
Evaluated models LLaMA-2-7B-Chat , Qwen2.5-7B-Instruct Abstract 作者称越狱来自两类头竞争，HCS 加蒸馏可把该竞争用于推理时干预。作者用注意力头机制分析 continuation-triggered jailbreak：同一有害指令下，把 continuation 后缀从指令后面挪到提示词边界外，模型更容易顺着后缀继续生成。
You have reached the end