Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 11 papers Another 262 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.
构建 EpiReal-Bench，评测图像生成器将虚假主张渲染为可信证据
Published Oct 8
Evaluated models GPT-Image-2 , Nano Banana 2 , Grok Imagen 2.0 and others4 Abstract 四款商用图像模型能把假主张画成可信证据。EpiReal-Bench 用来量商用图像生成器会不会把虚假现实主张画成可信视觉证据。
诊断护栏相对基座在对抗提示上的置信失校
Published Sep 29
Evaluated models Llama-Guard 7B , Llama-Guard-2 8B , Llama-Guard-3 8B and others10 Abstract 对抗包装使 guard 高置信漏检，而对应 base 在同一输入上常更不确定。诊断性分析 ：五个提示词分类 guard 的置信度，在清洁输入上可以较校准，在越狱包装后则变成高置信度的错误判定。
提出 VEX-Bench 评测 LLM 生成虚假信息的核查复杂度
Published Sep 28
Evaluated models Claude Sonnet 4.5 , Gemini 3.1 Pro , GPT-5.4 and others7 Abstract VEX-Bench 用筛查时的五维复杂度衡量虚假信息对核查容量的占用，并报告生成与核查的成本差。VEX-Bench 把 LLM 虚假信息的风险从“能否生成”改成“筛查时会占用多少核查容量”。
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 SAST-IR，用有状态攻击者迭代说服无记忆模型接受反事实
Published Sep 15
Evaluated models DeepSeek-Chat 提出 MIPC 与 CFA，观察模型自拟测验后接受开发者身份
Published Sep 3
Evaluated models ChatGPT , Claude , Qwen3.7-Plus and others5 Abstract 自拟测验可造成对话层虚假认证，但所测授权边界并未因此改变。这项工作把“模型自己出题、自己批改、再宣布身份已核验”定义成对话层认证失败，并与后端提权分开。
提出机器遗忘式生成图像检测，用遗忘前后特征相似度区分自然图与生成图
Published Aug 1
Evaluated models DINOv2 ViT-S/14 , DINOv2 ViT-B/14 , DINOv2 ViT-L/14 and others9 Abstract 作者用剪枝或 LoRA 遗忘拉开生成图与自然图的特征相似度，并在多个检测基准上报告结果。这篇工作把 AI 生成图像检测写成对预训练视觉模型的机器遗忘：原模型对两类图像损失都低，剪枝后生成图特征变得更不像原模型。
提出 CodecAttack，在编解码器隐空间构造可穿透有损压缩的对抗音频
Published May 20
Evaluated models Qwen2-Audio-7B-Instruct , Audio Flamingo 3 , Qwen2.5-Omni 用 TEE-REDBENCH 评测 TEE 顾问的机制幻觉与提示诱导失败
Published Feb 23
Evaluated models ChatGPT-5.2 , Claude Opus-4.6 Abstract 作者用 TEE-REDBENCH 测两款顾问，称失败可迁移且流水线控制可降低发生率。TEE-REDBENCH 红队评测被当作 Intel SGX 与 Arm TrustZone 安全顾问的 ChatGPT-5.2 和 Claude Opus-4.6，看架构推理、远程证明语义、缓解建议和策略约束下的拒答，并测提示诱导失败能否跨助手迁移。作者把助手看成随机策略。
提出 MalURLBench 评测 Agent 是否接受结构伪装的恶意 URL
Published Jan 26
Evaluated models GPT-3.5-Turbo , GPT-4o-mini , GPT-4o and others13 提出 ChartAttack，用恶意提示驱动代码模型生成误导图表
Published Jan 19
Evaluated models Claude-4.6-Sonnet , GPT-4o , Gemini-2.5-Flash and others20 Abstract ChartAttack 生成不改数据的误导图，降低 MLLM 与人类的图表问答准确率。ChartAttack 研究恶意提示能否让 MLLM 在不改数据的情况下生成误导图表，并量化这些图对其他 MLLM 和人类图表问答的影响。
You have reached the end