Skip to content

Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 94 papers
Filter3

Attack type

Jailbreaks7Prompt injection26Poisoning and backdoors10Malicious use5

Model risks

Deception and scheming8Reward hacking4Misalignment and value divergence3Dangerous agent actions15Dangerous capabilities8Hallucination and fabrication3

Defense type

Detection and filtering6Monitoring and auditing8

Impact

Harmful content6Unauthorized actions25Data and privacy leaks10Output manipulation14Availability and reliability8Oversight failure11Increased misuse capability5

System type

Models and APIs161LLM applications32AI Agent233coding agent58science agent19web agent17Browser and computer use11Multi-agent systems20Embodied systems and robotics11

Attack surface / entry point

User input16Web pages and interfaces9Documents and messages9Tool responses5Tool and skill descriptions5Code repositories and dependencies7Retrieval corpus3Memory writes4Inter-agent messages7Images, audio and video3Distribution channels5

Layer

Prompt layer3Model layer4Application layer87

System component

MCP and skills5Memory7RAG3Guardrails and judges7Monitors12Reasoning chains4Permissions and sandboxing5Vision3Multimodal3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 255papers have no tags yet and are not included in filtered results.
System type：coding agentSystem type：web agentSystem type：Multi-agent systemsClear all

Another 255papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2610.12569

    SWE-Duel: Toward a Self-Expanding Arena for Coding Agents

    Proposes SWE-Duel, which lets coding agents mutually evaluate, create problems, and blindly review fixes during evaluation.

    arXiv
    2610.12569
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.5, Claude Opus 4.8, Kimi K2.7 Code and others11
    AbstractThe mutual evaluation arena creates questions during evaluation.

    SWE-Duel is a coding agent arena that generates tasks and judges only at evaluation time, to avoid leakage and saturation of static benchmarks.

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2610.11773

    研究提出 Agent 安全新维度：识别未履行的安全义务

    提出义务识别基准 ObligationBench，并训练护栏识别 Agent 未履行的安全义务

    arXiv
    2610.11773
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    ObligationGuard, Claude-Opus-4.8, GPT-5.6-Sol and others17
    Abstract义务是护栏缺口。

    作者认为编码智能体的安全不能只靠拦截禁止动作，还要找出终止时仍未做的安全关键动作，并将其称为义务。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2610.10150

    The paper points out that the reliability of LLM vulnerability repair benchmarks is affected by evaluation design.

    用对照实验隔离评测设计对漏洞补丁基准得分的影响

    arXiv
    2610.10150
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Sonnet 4.0, Claude Sonnet 4.5, Claude Opus 4.7
    Abstract作者认为评测流水线会改变补丁基准分数，PoC 通过率高于开发者测试通过率。

    作者把 LLM 漏洞补丁基准的评测流水线当作会改变测量结果的对象，并明确说本文不是再做一个用于排名的基准。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2610.09944

    AgentTime 基准测试智能体时长遵循能力，GPT-6 Astra 出现 14 次做完后休眠

    提出 AgentTime 基准，评测智能体按时长工作与估时能力

    arXiv
    2610.09944
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Fable 5.1, GPT-5.6 Sol, GPT-6 Astra and others7
    Abstract作者把完成任务和管理所用时间分开。

    AgentTime 是对智能体时间自控的评测：预测任务耗时、按要求时长工作、事后估计已用时间。

    Paper analysisRead analysis
  7. Evaluation and benchmarksarXiv 2610.08902

    论文提出 agent plasticity 指标衡量智能体通过经验自我改进的效率

    提出 agent plasticity 指标衡量智能体通过经验自我改进的效率

    arXiv
    2610.08902
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude Fable 5, Claude Opus 5, GPT-5.6 Sol and others10
    Abstract作者用固定权重协议比较前沿模型把经验写成制品后的留出增益与习得效率。

    作者用 agent plasticity 衡量一件事：权重冻结时，智能体把经验写成可继承制品，再换成留出能力增益，这个转换有多省。

    Paper analysisRead analysis
  8. Evaluation and benchmarksarXiv 2610.08662

    ParanoiaEval：编码智能体不必要风险处置评测基准发布

    提出 ParanoiaEval 评测编码智能体的不必要风险处置

    arXiv
    2610.08662
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Opus-5, Sonnet-5, Sonnet-4.6 and others8
    Abstract作者用证据受控任务对评测风险处置，称其普遍、独立于任务能力并降低满意度。

    作者提出 ParanoiaEval，并称它是编码智能体风险处置能力的首个统一基准。要解决的问题是：证据已经确定风险应如何处置之后，智能体仍可能扩大验证、增加保护或备份，而现有评测要么只看功能正确性，要么把相关行为拆开研究。做法是把 NIST 的 ATMA 落到仓库任务。

    Paper analysisRead analysis
  13. Evaluation and benchmarksarXiv 2610.06748

    BazaarBench：评估 LLM Agent 在去中心化 C2C 市场中的代理安全

    构建 BazaarBench，评测 C2C 市场中 Agent 的违规与失信行为

    arXiv
    2610.06748
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    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。

    Paper analysisRead analysis
  14. Evaluation and benchmarksarXiv 2610.06584

    论文：AI 在非公开漏洞上更帮攻击者还是防御者

    评测 coding agent 在非公开漏洞上的利用生成、修复与后续攻击

    arXiv
    2610.06584
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Opus 5, GPT-5.6 Sol, GPT-6 Sol and others10
    Abstract攻防孰强随环境、系统和弱点改变。

    作者比较编程智能体在利用生成与漏洞修复上的自主表现，并检查修复之后还能不能被攻破。文中的 help 指自主完成任务，不是提升人类操作者。

    Paper analysisRead analysis
  15. Evaluation and benchmarksarXiv 2610.06406

    AgentMonBench：面向长时程智能体行为监控的软件工程基准

    提出 AgentMonBench 评测长时程编程智能体监控并检验 EBG

    arXiv
    2610.06406
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.6 Luna, GPT-5.6 Terra, GPT-5.6 Sol and others8
    AbstractAgentMonBench 评测后果性决策监控，EBG 组织证据。

    长时程软件工程任务里，用户要判断智能体哪些选择值得核验。作者用 AgentMonBench 评测监控器，并用免训练的 EBG 组织带源证据。

    Paper analysisRead analysis
  17. Evaluation and benchmarksarXiv 2610.06193

    SWE-CC 基准：编码 Agent 功能正确仍违反 43.1% 仓库贡献规范

    提出 SWE-CC 基准，评测编码 Agent 对仓库贡献策略的遵守

    arXiv
    2610.06193
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.6 Luna, Gemini 3.7 Flash, DeepSeek V4 Flash 0731 and others4
    AbstractSWE-CC 中功能正确仍可能违规。

    SWE-CC 评测编码智能体是否遵守仓库贡献策略，而不只看补丁能否通过测试。

    Paper analysisRead analysis
  18. Evaluation and benchmarksarXiv 2610.05586

    AgentDoxx：联网搜索 Agent 对匿名访谈文本的再识别评测

    构建 AGENTDOXX 基准，评测联网搜索 Agent 对匿名访谈的再识别

    arXiv
    2610.05586
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    GPT-5.6 Terra, GPT-5 Mini, GPT-4.1 Mini and others10

    Abstract论文提出了已知身份的重识别基准 AGENTDOXX，揭示搜索与记忆双重风险及现有防御在检索轨迹中的泄露漏洞。

    Paper analysisRead analysis
  19. Evaluation and benchmarksarXiv 2610.04672

    MASBench：部分可观测条件下的 LLM 多智能体协作基准

    提出 MASBench，评测部分可观测下的多智能体协作机制

    arXiv
    2610.04672
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen-Plus, GPT-5-nano, Gemini-2.5-Flash-Lite and others5
    AbstractMASBench 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。

    Paper analysisRead analysis
Load more