Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 14 papers Another 232 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.
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。
提出条件特征抑制防御，检出对抗补丁后抑制 VLA 内部特征
Published Oct 2
Evaluated models π0.5 , SmolVLA Abstract 条件抑制与 patch 相关的 SAE 特征，间歇攻击下可提高 SR，持续攻击下则不能。
提出 OmniConfess，冻结候选并按通道重打分以缓解全模态幻觉
Published Oct 2
Evaluated models Qwen2.5-Omni-7B , Qwen3-Omni-30B-A3B , Nemotron-3-Nano-Omni-30B 提出训练无关的 SECRET，对问题表征做源条件转向以缓解音视频模型幻觉
Published Sep 29
Evaluated models VideoLLaMA2-AV-7B , Qwen2.5-Omni-7B , Qwen3-Omni-30B-A3B Abstract SECRET 在问题状态上做源条件转向，三个 AVLLM 上减轻了跨模态接地幻觉。SECRET 是一种训练无关的问题状态转向方法，用来减轻 AVLLM 的 source-confused grounding hallucination。
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 SPARK，在预填充阶段修复多模态 KV 记忆以防御视觉语言模型越狱
Published Sep 13
Evaluated models LLaVA-OneVision-7B , Chameleon-7B , Qwen2-VL-7B and others4 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。
提出 TRIM，在黑盒推理时定位并净化图像后门区域
Published Sep 3
Evaluated models ResNet-50 提出 INTENT-AS-A-TOOL，用意图工具概率跟踪并干预智能体失准
Published Aug 28
Evaluated models Gemma-4-31B-IT , Qwen3.5-27B , Qwen3-235B-A22B and others5 Abstract 用意图工具的下一动作概率跟踪推理中的失准承诺，并在其成为首选时插入反思。Zhang 等人研究推理型语言模型在职场邮件沙箱里如何走向智能体失准，并提出用动作偏好而不是事后文本标签来跟踪承诺。问题是：有害执行往往出现在推理已经形成意图之后，但 CoT 监控是事后、粗粒度、依赖外部评审的，也看不到模型倾向调用哪个动作。
提出 SPAR 免训练注意力干预，削弱多模态模型的结构偏置
Published Jul 27
Evaluated models LLaVA-1.5-7B , LLaVA-NeXT Abstract SPAR 以结构偏置净化缓解视觉幻觉，结果需区分误导后收益与原始任务例外。SPAR 是一种无需训练、直接修改多模态内部注意力的幻觉防御。 作者关注视觉信息被查询不变的注意力偏置稀释，以及误导文本进一步推动模型依赖语言先验的问题。
提出 Agent MechSuits，对多轮 CLI Agent 做步级检测与子空间转向
Published Jun 22
Evaluated models LLaMA-3.1-8B , Qwen-2.5-7B , Gemma-2-9B Abstract 白盒步级探针加稀疏转向可压低三模型恶意动作，同时提高生成崩溃率。Agent MechSuits 是面向多轮 CLI 智能体的白盒、步级表示防御，并配有可执行轨迹基准 MAS。
Proposes inference-time defense ICON, which detects attention over-focusing and redirects to correct indirect prompt injection in agents.
Published Feb 24
Evaluated models Qwen-3-8B , Llama-3.1-8B , Mistral-8B and others6 Abstract ICON detects over-focusing and redirects at inference time to balance IPI safety and task utility.ICON detects attention over-focusing from IPI at inference time and redirects, avoiding refusal that would terminate the agent trajectory.
提出推理时防御 SafeThink，监视思维链并在低于阈值时注入转向前缀
Published Feb 12
Evaluated models R1-Onevision-7B , OpenVLThinker-7B , VLAA-Thinker-7B and others6 You have reached the end