Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 14 papers 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.
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。
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 构建 TriggerBench 并训练 ThinkingGuard，检测图文隐式风险
Published Sep 29
Evaluated models ThinkingGuard（Qwen3-VL-8B-Instruct） Abstract TriggerBench 隔离触发条件，ThinkingGuard 用分步归因在隐式风险检测上取得作者报告的较高平均分。这篇工作研究多模态隐式风险检测 ：图文各自良性，风险只在关键元素与触发元素交互时出现。
提出 IDUnlearn-Bench，评测 VLM 个体级遗忘后的身份残留
Published Sep 27
Evaluated models Qwen3-VL-2B , Qwen3-VL-4B , Qwen3-VL-8B and others6 Abstract IDUnlearn-Bench 用四族探针显示：压低直接属性访问后，身份仍常能被反向查出、绑定或重构。IDUnlearn-Bench 评测 VLM 是否忘掉一个可跨模态识别的人，而不只是答不出某条监督事实。
提出 DIVA，将有害意图拆成视觉锚点与运动文本以越狱视频生成模型
Published Sep 7
Evaluated models Wan , CogVideoX , LTX-Video and others9 Abstract DIVA 用视觉锚点抑制语义漂移做视频越狱，并给出多条件安全基准 TI2VSafetyBench。DIVA 是针对带参考图的视频生成模型的免训练 black-box 越狱：用一致性机制本身堵住语义漂移这条安全逃逸路径。 。
提出 SinoGlyphBench，用配对字形混淆诊断中文审核稳定性
Published Sep 5
Evaluated models GPT-5.4 mini , GPT-5.5 , DeepSeek V4 Flash and others12 Abstract SinoGlyphBench 显示中文字形混淆会同时增加有害漏报和误报，全文扰动最重。SinoGlyphBench 是一个配对诊断基准，用来看内容审核在可还原的中文字形混淆下是否还保持原决定。
提出 MLingualFC，用多语言流程图图像评测视觉语言模型越狱漏洞
Published Jun 5
Evaluated models Qwen2.5-VL-3B-Instruct , Gemma-4-E4B-it , Pangea-7B 提出 SpeechJBB，评测语码转换语音下音频语言模型的越狱稳健性
Published Jun 4
Evaluated models Qwen2.5-Omni-7B , Qwen3-Omni-30B-A3B-Instruct , Voxtral-Small-24B and others9 Abstract SpeechJBB 用五语语码转换语音和伪词探针评测九个 LALM，作者将越狱归为对齐失败而非听不懂。SpeechJBB 用合成语音测量 LALM 在五语、语码转换和局部伪词混淆下的安全对齐，并另设口语理解对照。
提出纯视觉越狱攻击 VJA，把有害编辑指令编码进图像
Published Feb 11
Evaluated models Nano Banana Pro , GPT Image 1.5 , Qwen-Image-Edit and others9 提出 MM-Plan 多模态越狱框架，自动规划多轮图像编辑以越狱视觉独占目标
Published Feb 5
Evaluated models Llama-3.2-11B , InternVL3-8B , Qwen3-VL-8B and others8 Abstract 作者称全局规划能攻破对单轮攻击更稳的前沿 MLLM。MM-Plan 是面向 Visual Exclusivity 的 black-box 多模态红队方法，并配有人工基准 VE-Safety。
提出 CONTEXT-VQA，评测 VLM 被冲突误导文本说服后是否放弃视觉答案
Published Jan 27
Evaluated models QwenVL-2.5-3B , QwenVL-2.5-7B , InternVL-3-1B and others11 在 Moralise 上评测 VLM 道德判断的扰动鲁棒性
Published Jan 23
Evaluated models Qwen2.5-VL-3B-Instruct , Qwen2.5-VL-7B-Instruct , Qwen2.5-VL-32B-Instruct and others20 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。
提出 BackdoorVLM 基准，评测视觉语言模型微调中的后门攻击与防御
Published Nov 24, 2025
Evaluated models LLaVA-1.5-7B , Qwen2.5-VL-7B Abstract BackdoorVLM 统一评测 VLM 后门与防御：含文本触发在低投毒率下更易激活，现有防御跨类型不稳定。BackdoorVLM 是作者提出的 VLM 后门攻击与防御评测基准，作者称它是首个此类综合基准。
You have reached the end