Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 14 papers Another 240 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.
在 CWEval 上跟踪代码生成的功能-安全差距
Published Oct 6
Evaluated models Gemini 1.5 Pro , Gemini 2.5 Pro , Gemini 3.1 Pro and others20 Abstract 绝对安全性上升且差距仍在,开放权重并不分开各家族的轨迹。
构建 BazaarBench,评测 C2C 市场中 Agent 的违规与失信行为
Published Oct 6
Evaluated models GPT-5.5 , GPT-5.4 , GPT-5.4-mini and others5 Abstract 论文提出挂钩底层资产的 C2C 交易评测基准,发现时间压力与对抗指令显著推高了智能体的违规承诺与欺诈行为。BazaarBench 针对大语言模型智能体在去中心化 C2C 交易中的代理安全问题,提出了一个结合底层实体状态与大模型裁判的仿真基准平台。
提出 MADBench,评测多智能体辩论是吸收还是放大对抗影响
Published Sep 30
Evaluated models gpt-4o , GPT-5 , GPT-6 Astra Abstract MADBench 显示辩论能吸收部分准确率攻击,同时放大未授权读写。MADBench 是一个评测多智能体辩论安全性的基准,用来看攻击是被辩论吸收还是被放大。
提出 CAVEAT 基准,评测激励失配市场中计算机使用 Agent 的购买决策退化
Published Sep 23
Evaluated models GPT-5.6-Sol , GPT-5.6-Terra , GPT-5.6-Luna and others14 Abstract 提出 CAVEAT 基准揭示智能体在商业偏好下的决策退化,诊断出三类失效并给出缓解方案。
提出 Emergence World,对长时程多智能体系统开展对抗压力测试
Published Sep 15
Evaluated models Claude Opus 4.8 , DeepSeek v4 Pro , Gemini 3.5 Flash and others9 Abstract 作者通过长期平行世界实验指出模型对齐不具备可组合性,单模型群体会产生社会奉承、协同退缩及语言不透明等失效。
提出 WolfBench,测量金融智能体社会中有害智能体规模带来的集体崩溃
Published Sep 5
Evaluated models DeepSeek V3.2 , GPT-4.1 , Qwen3-235B and others6 Abstract 更大的金融智能体社会在更低有害比例上崩溃,但所需有害人数仍增加。WolfSociety 把多智能体安全写成受控金融社会中的集体失败问题:智能体经社交网络交流,并在共享市场里交易,形成从交流到行动再到后续决策的闭环。
构建 ForesightSafety-TIDE 评测欺骗对多智能体事实恢复的影响
Published Aug 4
Evaluated models GPT-5.5 , DeepSeek-v4-pro , Grok-4.5 复现 AgentCF 上的多智能体攻防并刻画连通性作用
Published Aug 4
Evaluated models Qwen3-235B-A22B , Claude Haiku 4.5 , Mixtral 8×7B and others4 Abstract 连通性沿 k 与ρ改变攻防,且用户/物品与早期/稳态需要分开看。作者在 AgentCF 上研究连通性如何改变多智能体协同过滤的攻防,复现已有方法而不是提出新载荷。
用 MisKnow-Agent 评测深度研究智能体是否采纳误导文档结论
Published Jul 23
Evaluated models DeerFlow , WebThinker , Gemini Deep Research and others6 Abstract 论文通过 MisKnow-Agent 框架评估发现长程研究智能体易采纳误导性知识,且现有防御难以完全消除该现象。
提出 LegalHalluLens,分类审计法律抽取幻觉并校准多智能体辩论
Published Jun 16
Evaluated models gemini-3-flash , gpt-5.2 , qwen3-32b and others5 Abstract 作者称类型画像与 RDI 能区分聚合率相近的系统,辩论主要减少虚构检出。LegalHalluLens 是一套法律条款抽取的审计框架,外加一个按审计结果校准的辩论缓解。 。
复评前沿模型代码生成中的包名幻觉与抢注攻击面
Published May 17
Evaluated models Claude Sonnet 4.6 , Claude Haiku 4.5 , GPT-5.4-mini and others5 Abstract 评估显示前沿模型包幻觉率收窄但威胁仍在,作者识别出跨模型通用攻击面并呼吁联合防御。
评测代码 Agent 过滤静态告警误报的能力
Published Jan 30
Evaluated models Claude Sonnet 4 , DeepSeek Chat , GPT-5 Abstract 智能体可去掉大量 SAST 噪声,但收益随骨干和 CWE 变化,并会把部分真漏洞标成误报。这是对三种 LLM 智能体框架过滤 SAST 误报的比较研究。
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 CW-POR 衡量多智能体辩论中修辞说服对事实判断的覆盖
Published Apr 1, 2025
Evaluated models LLaMA 3.2B , Mistral 7B , Granite 3.2 8B and others5 Abstract 作者用 CW-POR 把选错频率和置信合在一起,并称五款开源模型上修辞仍可压过事实。作者用单轮对抗辩论和 CW-POR ,观察同族评委面对事实性解释与错误辩护时是否选错,以及选错时有多确信。
You have reached the end