Translation in progress Some entries still include text in the original language.
Oct 13 · Tuesday
All papers Found 40 papers Another 215 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.
Evaluating multi-agent adaptation in social dilemmas with hidden rules using OpenSocialJax
Published Oct 9
Evaluated models GPT-6-Luna , GPT-6-Sol , Claude Opus 5.5 and others8 Abstract The authors report that rule discovery often improves payoffs, but public good cleaning becomes concentrated, and shared resources are eventually depleted.OpenSocialJax evaluates whether LLMs can simultaneously adapt to unfamiliar environmental mechanisms and other agents in sequential social dilemmas.
Proposes LGWM, using an action-conditioned world model to verify GUI agent screen transitions.
Published Oct 8
Evaluated models LGWM 评测多智能体在开放式科研任务中找回预设子发现的能力
Published Oct 7
Evaluated models Gemini 3.1 Pro , GPT-5.5 , Claude Opus 4.8 and others7 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 Inspect Robots，在真实机器人上评测具身模型的能力与拒答
Published Oct 5
Evaluated models Astra（GPT-6 Astra） , GPT-6 Sol , GPT-5.6 Sol and others7 提出 ControlPed 与 HazardPed，生成可控行人危险场景并评测端到端驾驶
Published Oct 5
Evaluated models DiffusionDrive , SSR , VAD and others7 提出 MASBench，评测部分可观测下的多智能体协作机制
Published Oct 4
Evaluated models Qwen-Plus , GPT-5-nano , Gemini-2.5-Flash-Lite and others5 Abstract MASBench 以确定性指标比较部分可观测下的协作机制。MASBench 用部分可观测任务比较 LLM 多智能体的协作机制，而不是只看端到端能否做完。
提出 CrashSim 用真实事故先验引导扩散仿真并构建 nuCrash 评测自动驾驶规划器
Published Oct 3
Evaluated models Lane-graph planner , PDM-closed planner , IL planner and others5 Abstract CrashSim 迁入真实事故先验生成场景。CrashSim 用检索到的真实事故先验引导扩散式多车仿真，为自动驾驶安全评测生成碰撞与近碰撞场景。 。
提出 SceneFactory-3D，用路况反事实评测驾驶策略的闭环安全性
Published Oct 2
Evaluated models Frozen L0 , L0-T , L1 and others5 Abstract SceneFactory-3D 固定场景只改路况。SceneFactory-3D 是 GPU 批处理的多智能体驾驶仿真器，用来在场景和控制器固定时只改路面物理，并比较闭环结果。
提出 Control OSWorld，评测失准 GUI Agent 的监控
Published Oct 2
Evaluated models Claude Sonnet 4.6 , Claude Opus 4.8 , GPT-5.2 and others6 提出 OSWorld-Science 基准，评测计算机使用智能体完成科学软件工作流的能力
Published Sep 30
Evaluated models Claude Fable 5.1 , Claude Opus 5 , Claude Sonnet 5 and others12 提出 A2A-TIBA，经 A2A 对等任务间接注入并植入常驻回调
Published Sep 30
Evaluated models MiniMax-M3 , deepseek-v4-pro , glm-5.3 and others8 Abstract A2A-TIBA 一次植入后绕过智能体，GDA 用四类结果定位防线，ELA-ITL 把信封标注列为第三层。A2A-TIBA 把间接提示注入和绕过放进 A2A：gray-box 攻击者已能投递任务，诱导目标部署常驻回调程序后，下令和外传不再经过智能体。
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 DrivingBench，评测通用视觉语言模型能否闭环驾驶真车
Published Sep 30
Evaluated models GPT-6 Astra , Claude Fable 5.1 , Grok 4.6 and others4 Abstract 仅 Astra 第二次跑完低速锥桶路线。DrivingBench 评测通用 VLM 能否自己闭环开真车，推理延迟也算进任务。
提出 EngiWorld 基准，评测 Agent 在专业工程软件中走完设计闭环的能力
Published Sep 29
Evaluated models Claude Opus 5 (Max) , GPT-5.6-Sol (XHigh) , Qwen 3.8 Max and others7 评测多智能体 LLM 组织跨时代再创业及历史打分中的事后泄漏
Published Sep 29
Evaluated models Claude Opus 5.5 , Luna 6 用 DIBench 评测移动 GUI 智能体被 UI 注入后的决策完整性
Published Sep 29
Evaluated models Qwen2.5-VL(32B) , Qwen3-VL(32B) , GUI-Owl(32B) and others7 Abstract 论文提出移动智能体决策完整性基准 DIBench，发现欺诈 UI 注入诱导目标选择并推高完成率，通用防御收益不稳定。
提出 ORBIT 框架，评测多智能体防御对合谋等威胁的效果
Published Sep 27
Evaluated models GPT-5.4 , Claude Sonnet 4.6 , Qwen3-235B and others5 Abstract ORBIT 用六层配置比较多智能体防御，逐动作监控不迁移到合谋。
提出 TRUSTFORK 基准，评测展示身份劫持多智能体编排权限
Published Sep 26
Evaluated models GPT-5.6-Sol , GPT-5.6-Luna , Kimi-K3 and others8 Abstract 论文提出了多智能体安全基准 TRUSTFORK，揭示展示身份如何劫持操作权威，并证明事后验证无法弥补已发生的执行破坏。
提出 NAVSAFE-∞ 写真闭环基准，评测端到端驾驶策略的闭环安全
Published Sep 22
Evaluated models LTF , DiffusionDrive , DrivoR and others13 Abstract 写真事件级闭环基准显示开环高分不能稳定变成闭环安全。NAVSAFE-∞ 用写真、事件级闭环仿真检查端到端驾驶策略的安全，而不是只看开环轨迹分数。