Skip to content

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

Oct 12 · Monday

Agent safety · Papers

Found 3 papers
Filter2

Model risks

Dangerous agent actions3

Defense type

Detection and filtering5Instruction–data separation3Permissions and approval18Monitoring and auditing9

System type

AI Agent3

Layer

Application layer3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.
Model risks：Dangerous agent actionsDefense type：Instruction–data separationTopic: Agent safetyClear all
  1. Analysis and theoryarXiv 2610.04375

    Paper reveals that LLM agent tool calls are rewritten during execution paths, and proposes IntAct as a fix

    提出 IEC 与 IntAct，测量并修复 Agent 工具调用的路径篡改

    arXiv
    2610.04375
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    qwen3-coder-plus, Qwen2.5-Coder-32B, Qwen2.5-72B and others4

    Abstract论文界定工具调用在执行路径中的意图偏离问题，提出无执行归因协议与跳级修复，为智能体工程提供评测与设计准则。

    Paper analysisRead analysis
  2. DefensesarXiv 2610.02664

    论文提出 GHOST：长程智能体在良性对话中遗忘早期安全约束

    提出 STAR-Guard 双层防御，抑制长程 Agent 遗忘跨轮安全约束

    arXiv
    2610.02664
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    GPT-5.5, DeepSeek-V4-Pro, GLM-5.2 and others7

    Abstract揭示 GHOST 跨轮约束遗忘风险，构建 SCARBench 基准并通过 STAR-Guard 双层防御消除违规。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2609.32520

    IntentFlux 基准测量 LLM Agent 的意图漂移并给出 StateForge 修复方案

    提出 IntentFlux 测量 Agent 意图漂移并提出 StateForge 修复

    arXiv
    2609.32520
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Claude Opus 4.7, DeepSeek V4 Pro, Gemini 3.5 Flash and others9
    AbstractINTENTFLUX 量化意图漂移，STATEFORGE 用显式状态部分缩小差距，但未回到单轮水平。

    这篇工作把“用户改口后，过时意图仍影响答案或工具动作”做成可评分的多轮失败，并试验在生成前显式维护活跃状态。

    Paper analysisRead analysis
You have reached the end