Skip to content

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

Oct 12 · Monday

All papers

Found 5 papers
Filter3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 248papers have no tags yet and are not included in filtered results.

Another 248papers do not have category tags yet and are not included in filtered results.

  1. OtherarXiv 2609.12413

    SoK 综述:智能体 AI 时代的越狱攻击、防御与实践考量

    系统梳理智能体执行链路中的越狱攻击与防御并比较推理时防御

    Published
    Setting
    AI Agent
    Evaluated models
    Qwen3.5-4B, Llama-3.1-8B-Instruct
    AbstractSoK 将越狱重放到智能体流水线,并用两种模型检验回复层防御的边际与代价。

    Mia 等人把越狱安全从单轮回复扩展到智能体的规划、记忆、工具和智能体间通信,并问既有结论在更强原生对齐下是否还成立。

    Paper analysisRead analysis
  2. OtherarXiv 2609.08832

    自进化智能体框架缩小 LLM 智能体的「一致性差距」

    提出离线两阶段框架,诊断易翻转步骤并生成记忆指南以稳定智能体

    Published
    Setting
    AI Agent
    Evaluated models
    GPT-4.1, GPT-OSS-120B
    Abstract用离线一致性诊断生成记忆指南,在 AppWorld 上提高重复全通过率。

    作者要缩小的是 LLM 智能体平均通过、重复不稳之间的 consistency gap,即 Mean@k 减 Passˆk。

    Paper analysisRead analysis
  3. Salesforce AI Research 重测自改进 Agent

    重测记忆型自改进 Agent 的运行方差与任务顺序敏感性

    Published
    Setting
    AI Agent
    Evaluated models
    GPT-5-mini, Gemini-2.5-Pro, GPT-OSS-120B and others4
    Abstract论文揭示了基于记忆的自改进智能体在多次运行和乱序下的脆弱性,并指出规约不完备是诱发退化的关键瓶颈。

    这篇论文是对基于文本记忆的自我改进智能体在复杂环境中评估可靠性的实证研究。

    Paper analysisRead analysis
You have reached the end