Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 39 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.
Propose KVeil, which adds perturbations to face videos to prevent unauthorized customization from reproducing identity
Published Oct 9
Evaluated models Wan2.2-TI2V-5B , HunyuanVideo-1.5 , Kling 3.0 and others4 Abstract KVeil adapts LETHE's reference-reading disruption to the challenge's two stages, ranking first on the leaderboard in both Final columns.
提出校准重合成方法,以生成器反演保真度认证真实图像并防御对抗扰动
Published Oct 5 Abstract 论文提出基于反演保真度与阈值校准的健全认证框架,在对抗扰动下维持错误率界限,并指出真实内容可认证性随生成器演进而萎缩。
提出 TRACER,从权重差识别扩散模型被擦除概念
Published Oct 5
Evaluated models Stable Diffusion v1.5 , Stable Diffusion 3 , FLUX.1 and others5 Abstract TRACER 用权重足迹识别未披露擦除概念并估计个数,作者称数秒内完成且准确率高于对照。TRACER 针对的是遗忘扩散模型上的目标识别:提供者从公开基座 W 用已发表方法擦掉未知集合 S⋆,只发布 W′。攻击者还有假定包含这些概念的词表,但不知擦了谁、擦了几个、用了什么算法。
提出 LiBRA,在潜空间双向优化以规避图像水印双侧检测
Published Oct 2
Evaluated models Stable Signature , YU1 , HiDDeN and others4 Abstract LiBRA 用双向潜空间优化规避双侧水印检测,并提高相对反转基线的保真。
提出 UNTAG,绕过开源权重模型的 trigger-tag 滥用检测
Published Oct 2
Evaluated models Llama2 (7B) , Llama3 (8B) , Qwen2.5 (7B) Abstract 作者用 UNTAG 评估 trigger-tag,认为其不能充当开放权重滥用检测的主要边界。UNTAG 是针对开放权重 LLM 上条件滥用检测 trigger-tag 的统一攻击评估:信号可以在解码期注入,也可以写进权重,但攻击者若能改输出或改已发布权重,检测可以失效。
提出 ReGap,用自生成伪监督微调恢复模型中的隐私关联
Published Oct 1
Evaluated models GPT-2 Small , GPT-2 Medium , GPT-2 Large and others6 Abstract 提出无真实私有监督的恢复攻击 ReGap,实验显示微调可通过自生成数据唤醒模型潜在隐私关联,突显后适应隐私风险。
提出 Whiteout,用伪装样本微调覆盖个人敏感信息关联
Published Oct 1
Evaluated models Llama 3.2 , Phi-3-mini , Gemma 3 and others7 提出 GSU 在发布前封堵敏感门梯度,抵抗下游微调获取禁止能力
Published Sep 30
Evaluated models Llama-3.2-1B , Llama-3.2-3B , Llama-3.1-8B and others7 Abstract GSU 在发布前封堵 SiLU 门的获取梯度,并在 TOFU 与 WMDP 上降低下游微调得分。Gradient-Sealed Unlearning(GSU) 是一种发布前的 preemptive unlearning:不删除模型里已经有的能力,而是让指定禁止域能力更难被发布后的微调获取。
提出 PACT,遗忘推理模型在压力下的欺骗行为并保留上下文理解
Published Sep 30
Evaluated models DeepSeek-R1-Distill-Qwen-32B , QwQ-32B Abstract 论文将大模型欺骗形式化为上下文条件行为,提出 PACT 框架在消除欺骗的同时保留了上下文遵循能力。
检验层冻结与奇异方向截断能否挡住少样本有害微调
Published Sep 29
Evaluated models Llama-3.1-8B-Instruct , Llama-3.1-Tulu-3-8B-DPO , OLMo-2-1124-7B-Instruct and others10 Abstract 定位与恢复可以复现,但知情攻击能把损伤移出冻结带,并打败跨检查点的截断修复。这篇工作检验:把拒答定位到某些层或更新方向之后,能不能在攻击者知情时还守住拒答。 。
提出跨架构攻击 CSR,用概念分数参数重学习恢复文生图被擦除概念
Published Sep 27
Evaluated models FLUX.1-dev , Stable Diffusion-v1.4 提出 AEGIS,用中层内部信号选择性激活后层适配器拦截音频越狱
Published Sep 24
Evaluated models Qwen2-Audio-7B , VITA-1.5 , Ultravox v0.5 and others6 提出 PALS,用潜空间平滑防御全双工语音对话的对抗扰动
Published Sep 23
Evaluated models Moshi-style agent 提出 BoundStyle 与 StyleAT,对抗训练防御人脸识别语义攻击
Published Sep 20
Evaluated models MobileFaceNet (MobileFace) , ResNet (ResNet-152, IR-SE) , RepVGG and others7 Abstract BoundStyle 用 StyleGAN3 有界潜编辑快速攻击 FR。StyleAT 用可调的 StyleGAN3 潜空间攻击 BoundStyle 对人脸识别做对抗训练,以抵抗一般语义编辑,并在评测中面对训练时未见的 DiffPrivate。
提出 ARIA,在冻结权重上用 SAE 做推理时遗忘门控
Published Sep 15
Evaluated models DeepSeek-R1-Distill-Qwen-1.5B , Gemma-3-1B-it 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 提出 OBELISK 三级流水线,权衡恶意软件检测的准确率、开销与鲁棒性
Published Sep 8
Evaluated models GBDT-ALL , GBDT-YARA , NEBULA-ALL and others4 Abstract 三级过滤以不大的检测损失换速度,同一机制扩大攻击面,部署应按约束选配置。OBELISK 把 Windows PE 恶意软件检测做成 YARA、EMBER-GBDT 与 Nebula 的顺序流水线,用来同时看检出、计算开销和对抗鲁棒性。
揭示拆分训练返回梯度以零支撑暴露诱饵行并泄露训练内容
Published Sep 4
Evaluated models Qwen3-0.6B , 35B-A3B Abstract 论文揭示拆分训练中反向梯度零支撑完全暴露诱饵行,联合视图突破门控,裁剪加噪可有效缓解。这是一项关于双节点大语言模型拆分训练系统的系统安全案例研究,揭示了评估工具遗漏反向信道所导致的静默隐私失效。
提出 SEAL,在 MoE 共享专家上做 DPO-LoRA 对齐以抵御越狱与权重篡改
Published Sep 2
Evaluated models Qwen1.5-MoE-A2.7B , DeepSeekMoE-16B , GLM-4.7-Flash and others4 Abstract SEAL 把 DPO-LoRA 限制在混合 MoE 共享专家上,SEAL++再保护已有安全神经元方向。SEAL 是面向混合 MoE 的训练时防御:只对齐始终执行的共享专家,使安全不依赖路由器是否把 token 送到安全专家。
提出 DIVA,利用跨步条件传播对离散扩散视觉语言模型实施视觉越狱
Published Sep 2
Evaluated models LLaDA-V , MMaDA , LLaDA2.0-Uni