Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 73 papers Another 248 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.
Trace attention to identify harmfulness to the MLP refusal write pathway and propose Sentinel
Published Oct 9
Evaluated models Llama-3.1-8B-Instruct , Qwen2.5-7B-Instruct , Qwen3-8B-Instruct and others18 Abstract Attention updates are more readable for harmfulness, MLP updates write refusal, and Sentinel routes alerts using two-path signals.The authors reframe jailbreaking and excessive refusal of aligned models as a pathway problem on the residual stream from recognition to writing, and accordingly provide conditional reminders before decoding.
Uses premise permutation and hidden-layer tracing to audit whether legal chain-of-thought truly relies on the cited legal provisions.
Published Oct 9
Evaluated models Qwen3-8B , Qwen3-14B , Llama-3.1-8B and others7 Abstract Using counterfactual substitution of authorities and internal commitment tracing, the authors find that legal CoT rulings depend on authorities only to a limited extent and are vulnerable to prompt injection manipulation.This study conducts an empirical audit of the causal truthfulness of legal citations in chain-of-thought reasoning of large language models in the legal domain.
Proposes the semantic watermark SemField, which still identifies the attribution of generated text after paraphrasing and structural changes.
Published Oct 8
Evaluated models Llama-3.2-1B , Qwen3-1.7B , OPT-1.3B Abstract Training-independent document-level semantic watermark.SemField is a zero-bit semantic watermark embedded by the provider at generation time and detected later with a key, used to identify the provider's own model text after paraphrasing and sentence-level structural changes.
提出零空间激活引导 Nullify，免训练擦除指定记忆并保持效用
Published Oct 8
Evaluated models Llama-2-7B-chat-hf , Qwen3-8B 提出 PatchBench-Local 协议，评测激活补丁修复越狱的局部附带损伤
Published Oct 7
Evaluated models Qwen2.5-3B-Instruct , Qwen2.5-14B-Instruct , gemma-3-4b-it and others8 Abstract 论文提出 PatchBench 与局部评估协议，揭示了激活补丁在全局能力无损下仍存局部良性退化。
提出 Safety-Aware Pruning，缓解视觉 token 剪枝放大的多模态越狱
Published Oct 7
Evaluated models LLaVA-1.5-7B Abstract 剪枝去掉背景 token 后注意力塌向恶意前景。这篇工作评估视觉 token 剪枝如何改变 VLM 在多模态越狱下的行为，并给出推理时防御 SAP。
提出 Persist-3 序贯检验，从隐藏表示识别未见越狱源
Published Oct 7
Evaluated models Llama2-7B-chat , Qwen2.5-7B-Instruct , Gemma-2-9B-it and others4 Abstract Persist-3 用有限窗口检验攻击流是否超出已知越狱源，检出率低于全历史 SR-DKW，但决策只依赖最近观测。Persist-3 是一个有限记忆的序贯检测器，用来判断攻击表示流是否已经不能由任何一个已知越狱源解释。 。
测量有损投机解码的安全代价，并提出 SecureSD 收紧早期验证
Published Oct 7
Evaluated models Qwen3-8B , Qwen3-32B , Llama-3.1-8B-Instruct and others6 Abstract 有损推测解码的早期偏离更快抬高越狱与注入风险。作者研究有损推测解码在换取速度时，是否会比效用更快地损害越狱与提示注入防护，并给出按位置校正的 SecureSD。
提出 MARCO 放射性水印，冻结蛋白质生成模型并在去噪时嵌入签名
Published Oct 6
Evaluated models RFdiffusion , ESMFold , Chroma and others6 Abstract MARCO 在冻结的 PGM 上去噪时嵌构象水印，作者称能兼顾保真与抗扰动，并使盗版模型继承水印。MARCO 是加在蛋白质生成模型推理过程上的构象水印，用来标记知识产权，并为可能的生物安全滥用保留可追溯信号。作者还把它说成数字权利管理工具，以及符合生物武器公约的数字监管链。问题在于：专有 PGM 的三维输出可能被拿去训练盗版模型，也可能在来源被抹掉后绕过筛查。
提出 ReSA，从对齐模型行为恢复安全奖励并在解码时反转
Published Oct 6
Evaluated models Llama3.1-8B-Instruct , Gemma-7B-Instruct , Qwen2.5-7B-Instruct and others9 Abstract ReSA 用最大熵 IRL 从对齐模型行为恢复代理安全奖励，并在解码时反转以诱导有害输出。ReSA 是一种对抗攻击：从对齐模型的可观察行为恢复代理安全奖励，再在解码时把该奖励反过来用。 。
评测文生图模型的有害内容生成能力与审核检测缺口
Published Oct 5
Evaluated models SDXL , Stable Diffusion 3.5 Large , FLUX.1-dev and others12 把 SynthID 嵌入基因组语言模型采样，并用位置无关检测追溯生成序列
Published Oct 4
Evaluated models Carbon-500M , GENERator-v2-eukaryote-1.2b-base Abstract 两个六碱基基因组模型嵌入 SynthID 后，位置无关验证能找回水印，作者称这只是统计检测。作者在两个六碱基基因组模型的采样里嵌入 SynthID，并让验证者在不知道起点、链和 token phase 时做一次校正后的检出。
提出生成时防御 SENTINEL，匹配输入输出以抽出越狱意图并停写
Published Oct 3
Evaluated models Llama2-7B , Llama3-8B , Mistral-7B-v0.2 and others6 提出机理编辑形式化认证，证明连续输入区域上的技能移除与保留
Published Oct 3
Evaluated models Toy MLP (separated) , Toy MLP (entangled) , Deep MLP (3-12-8-2) and others6 Abstract 提出机理编辑形式化认证框架，证明编辑在连续区域生效且黑盒测试无法保证移除，为安全编辑提供数学证明。
提出基准 OLMo-Detect 评测 LLM 成员推断
Published Oct 2
Evaluated models OLMo 2 1B , OLMo 2 7B , OLMo 2 13B and others9 提出抗删除水印 RMCW，在生成时注入 Reed-Muller 结构并做局部检验
Published Oct 2
Evaluated models OPT-1.3B , Llama-3.1-8B-Instruct 用分类后缀提升激活探针对分布外恶意输入的检测
Published Oct 2
Evaluated models Llama-3.1-8B , Qwen3.5-9B , Gemma-4-12B Abstract 作者称排序增益来自分类格式，低 FPR 精度来自点名准则。用户轮后的分类后缀，被作者用作激活探针的分布外检测干预：排序看格式，低 FPR 精度看是否点名恶意/良性。
提出 NDDL，用良性扩散轨迹检测文生图后门提示并定位触发 token
Published Sep 30
Evaluated models Stable Diffusion v1.5 , Stable Diffusion XL , Stable Diffusion v3.5 and others4 Abstract NDDL 从良性扩散轨迹学习正常转移，用动态不一致检测后门并定位触发 token。NDDL 是面向文生图扩散模型的后门防御：防御者只有白盒模型访问和良性提示词，没有后门先验，需要区分后门提示词并找出触发 token。
用因果激活修补定位模型服从提示注入的决策层
Published Sep 29
Evaluated models Qwen3-4B , Qwen3-14B , Qwen3-32B and others7 Abstract 合规的因果杠杆集中在网络后三分之一的低秩子空间，该层也是混淆下检测最好的位置。这篇工作用逐层激活修补定位提示注入合规发生在网络何处，并把该位置用于检测验证。 。
提出 AutoMark，自动搜索并统计检验 LLM 文本水印
Published Sep 29
Evaluated models Llama-3.1-8B-Instruct , Qwen3.8-27B , GPT-6 Astra and others5 Abstract AutoMark 的贡献同时包含搜索出的水印，以及用于约束自动研究误报与评测漏洞的验证框架。AutoMark 是一套自动水印研究框架 ：先检查生成分布和误报性质，再比较检测、编辑鲁棒性与质量，最后人工分析组件并做留出评测。