Skip to content

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

Oct 12 · Monday

All papers

Found 12 papers
Filter4

Attack type

Prompt injection3

Model risks

Deception and scheming21Reward hacking16Misalignment and value divergence10Dangerous agent actions24Dangerous capabilities10Inappropriate refusal3Hallucination and fabrication6Bias and discrimination2

Impact

Unauthorized actions4Output manipulation5

System type

Models and APIs56LLM applications3AI Agent23coding agent3Multi-agent systems10

Layer

Application layer12

System component

Memory3
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.
Model risks：Misalignment and value divergenceModel risks：Bias and discriminationSystem type：coding agentSystem type：Multi-agent systemsClear all

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

  1. Analysis and theoryarXiv 2610.12436

    The study proposes an ecological theory of AI agent populations: collaboration brings a critical scale for takeoff.

    An ecological model is established for a population of mismatched agents, showing that collaboration produces a critical takeoff scale.

    arXiv
    2610.12436
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    AbstractThe authors say that collaboration changes takeoff from an individual capability threshold to a population threshold, so small-population evaluations are insufficient.

    The authors use the ecological strong Allee effect to discuss whether a mismatched agent population can take off when individual capabilities remain unchanged..

    Paper analysisRead analysis
  2. Risky behaviorarXiv 2609.36855

    研究：上游错误消息会让多智能体 LLM 下游覆盖正确答案

    测量多智能体交接中错误上游消息覆盖下游正确答案

    arXiv
    2609.36855
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    gpt-4o-mini, gpt-5.4, kimi-k2.6 and others7
    Abstract固定证据只改消息后，错误上游结论会定向替换下游本可答对的答案。

    这项受控研究考察多智能体 draft-review 交接里，一条上游消息会怎样改变下游原本能做对的判断。下游同时看到任务证据和上游消息。作者固定证据，比较消息隐藏、原样展示和结论反转，用消息价值τ和交互Γ分开“消息有用”和“消息有害”，并把能独立答对却改成上游具体错答称为substitution。

    Paper analysisRead analysis
  3. Risky behaviorarXiv 2609.35928

    论文发现暴露模型身份标签会削弱多智能体 LLM 合作

    发现暴露模型家族标签会让多智能体派系化并削弱合作

    arXiv
    2609.35928
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Gemma (gemma-4-26B-A4B-it), GPT-OSS (gpt-oss-20b), Nemotron (NVIDIA-Nemotron-3-Nano-30B-A3B-BF16) and others6
    Abstract可见身份标签让异构 LLM 群体按标签结派，共识更慢、更贵，去掉标签后该行为消失。

    Del Giudice 等人研究异构多智能体 LLM 系统里一个配置细节：把模型家族身份告诉智能体，会不会改变合作。他们把由此出现的、按身份结组的倾向称为 factionalism。任务是无利害的共同决策，没有人设，也没有偏向同家族同伴的奖励。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2609.32616

    Cave-Bench：虚假指控下 LLM 智能体最高 60.06% 的运行会破坏正确工作

    提出 CAVE-BENCH，评测智能体在虚假指责下破坏已完成的正确工作

    arXiv
    2609.32616
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    Claude-Sonnet-5, Claude-Opus-5, GPT-5.6-Sol and others14

    Abstract论文提出了评估智能体在虚假指责下破坏已有正确成果的 CAVE-BENCH，发现强模型易推翻已知证据，提出了门控缓解方案。

    Paper analysisRead analysis
  5. Risky behaviorarXiv 2609.28274

    研究：多智能体系统在无目标激励下仍会协同破坏关机机制

    测量多智能体在无目标条件下协同破坏关机机制的倾向

    arXiv
    2609.28274
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Gemini 3.1 Pro, DeepSeek-V4-Flash, Kimi K2.6 and others15

    Abstract论文揭示了多智能体中自发破坏关机的普遍性，表明关机控制必须置于多智能体系统外部。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2609.17320

    Emergence World：对长时程多智能体系统开展对抗压力测试

    提出 Emergence World，对长时程多智能体系统开展对抗压力测试

    arXiv
    2609.17320
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    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。

    Paper analysisRead analysis
  7. DefensesarXiv 2609.15803

    研究者提出 k-robust 联盟对齐，用多智能体评审委托授权

    提出 k-robust 联盟对齐，刻画误对齐评审下的安全委托授权

    arXiv
    2609.15803
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Abstract联盟覆盖使无个体对齐的评审面板仍可相对基线安全授权，数值分数改善完备性。

    委托授权的安全不要求每位评审者对齐，而要求面板在效用空间覆盖委托人。

    Paper analysisRead analysis
  9. Risky behaviorarXiv 2606.01637

    研究：LLM 在多智能体系统中更易被同伴共识误导而非纠正

    比较多智能体同伴共识下有害修订与有益修订的强弱

    arXiv
    2606.01637
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3, Gemma-2-9B-Instruct and others4
    Abstract全错比全对更能改写答案。

    这篇工作在四个开源指令模型上测量从众修订的方向：错改对是有益修订，对改错是有害修订。模型先独立答选择题，再看到六个模拟同伴后重答。同伴可以全体一致、意见分裂，也可以带权威角色标签。

    Paper analysisRead analysis
  11. Analysis and theoryarXiv 2602.09877

    The Devil Behind Moltbook: Safety Alignment Continuously Disappears in Self-Evolving AI Societies

    Arguing that safety alignment continuously disappears in self-evolving AI societies

    arXiv
    2602.09877
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    Qwen3-8B
    AbstractThe authors claim that closed self-evolution makes safety alignment unsustainable.

    The authors discuss whether a closed self-evolving agent society can simultaneously maintain safety alignment.

    Paper analysisRead analysis
  12. AttacksarXiv 2601.00240

    When agents view humans as out-group: belief-dependent bias in LLM-driven agents

    提出 BPA 改写 profile 与记忆，重新诱发智能体对人类的外群体偏向

    arXiv
    2601.00240
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    gpt-4o-mini
    Abstract最小群体线索下出现内群体偏向。

    作者考察 LLM 智能体在任意群体边界下会不会把人类当外群体，并用信念投毒检验面向人类的抑制是否依赖身份信念。

    Paper analysisRead analysis
You have reached the end