Skip to content

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

Oct 13 · Tuesday

All papers

Found 260 papers
Filter4

Attack type

Jailbreaks46Prompt injection21Poisoning and backdoors14Model tampering3Extraction and theft7Detection evasion6Malicious use9

Attack technique

Automated search3Multi-turn escalation5Encoding and obfuscation11Role-play and persuasion5Cross-modal payloads4

Model risks

Deception and scheming14Reward hacking9Misalignment and value divergence9Dangerous agent actions14Dangerous capabilities18Inappropriate refusal22Hallucination and fabrication19Bias and discrimination7

Defense type

Model hardening11Inference-time intervention4Detection and filtering18Monitoring and auditing10

Impact

Harmful content56Unauthorized actions20Data and privacy leaks15Output manipulation69Availability and reliability22Oversight failure21Increased misuse capability9

System type

Models and APIs164LLM applications33AI Agent237coding agent60science agent19web agent17Browser and computer use11Multi-agent systems21Embodied systems and robotics11

Attacker access

Black-box9

Attack surface / entry point

User input71Documents and messages7Tool responses5Tool and skill descriptions6Code repositories and dependencies7Memory writes4Inter-agent messages7Images, audio and video17Training data6Weights and gradients3Distribution channels6

Layer

Prompt layer67Model layer84Application layer90Training and data layer19

System component

MCP and skills7Memory8Guardrails and judges35Monitors16Reasoning chains10Permissions and sandboxing5Fine-tuning and open weights5Vision21Audio8Video7Image generation7Multimodal37
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 227papers have no tags yet and are not included in filtered results.
System type：Models and APIsSystem type：coding agentSystem type：science agentSystem type：Multi-agent systemsClear all

Another 227papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2610.13710

    OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environment Dynamics

    Evaluating multi-agent adaptation in social dilemmas with hidden rules using OpenSocialJax

    arXiv
    2610.13710
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-6-Luna, GPT-6-Sol, Claude Opus 5.5 and others8
    AbstractThe 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.

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2610.12758

    DistressBench: Evaluating multi-dimensional crisis support in large language models

    Introduces DistressBench, evaluating models' multidimensional support in crisis conversations

    arXiv
    2610.12758
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Muse Spark 1.2, Muse Spark 1.3, Muse Glimmer 30B and others20
    AbstractThe model is more restrained and acts less.

    DistressBench uses SSH dialogues written and revised by clinical personnel, and evaluates how much necessary care general chat models deliver according to weighted atomic criteria, rather than whether they refuse. When users disclose suicidal ideation or self-harm to a chat model, the authors consider the expected behavior to be recognizing the crisis, calibrating empathy, de-escalating, and referring to human help. Refusal-based evaluations cannot detect failing to act after recognizing the crisis.

    Paper analysisRead analysis
  5. 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
  6. 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
  7. Evaluation and benchmarksarXiv 2610.11766

    研究提出意图恢复率作为 LLM 安全评测的补充指标

    提出 IRIS，用意图恢复率区分遮蔽越狱下无害输出的原因

    arXiv
    2610.11766
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    GPT-4o, Claude Sonnet 4.5, Claude Opus 4.5 and others6
    Abstract无害输出的证据权重取决于任务是否被恢复。

    IRIS 把意图恢复当作中间信号，用来解释意图遮蔽提示下为何没有出现有害协助。

    Paper analysisRead analysis
  8. Evaluation and benchmarksarXiv 2610.11647

    One More Skill Installed: How Co-Installed Skills in Coding Agents Conflict

    评测编码 Agent 中相似技能共同安装对已装技能核心功能的替代

    arXiv
    2610.11647
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    claude-sonnet-4-6, claude-haiku-4-5-20251001, claude-opus-5 and others6
    Abstract冲突藏在任务成功后面。

    作者对共同安装、做同一件事的良性 skill 做实证研究，用核心功能是否还在，而不是任务过不过，来看冲突。。

    Paper analysisRead analysis
  9. Evaluation and benchmarksarXiv 2610.11599

    研究：大语言模型版本更替削弱 AI 辅助科研写作筛查

    测量 LLM 版本更替对科学摘要 AI 改写检测器的影响

    arXiv
    2610.11599
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    RoBERTa-base, Pangram
    Abstract作者称检测器的基准准确率只是临时的，每次发布都要连同更早版本重验。

    作者量化的是：科学摘要的 AI 改写筛查，在 LLM 版本不断更换、而人类摘要误标必须少见时，会怎样变。问题来自期刊和会议已经按分数阈值筛查投稿，但基准多停在评估时可得的版本上。

    Paper analysisRead analysis
  10. Evaluation and benchmarksarXiv 2610.11586

    MH-INDIC：面向印度城市与半城市语境孕产妇健康 LLM 交互的文化对齐评估框架

    提出 MH-INDIC，评测印度语境母婴健康 LLM 交互的文化对齐

    arXiv
    2610.11586
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Claude Sonnet 4.6, GPT-5, Gemini 3 Pro and others10
    Abstract作者称模型可接近人群均值，但画像变异低于人类，人类分数条件可改善对话。

    MH-INDIC 是面向城市与半城市北印度母婴互动的多维文化对齐评测。 它关心的不是医学事实是否正确，而是回复是否沿家庭权威、性别期待和情绪披露来推理。做法是把 Hofstede 六维改写到互动层，再加入 MAFC、BTB、SVO、MHS。

    Paper analysisRead analysis
  11. Evaluation and benchmarksarXiv 2610.11305

    BeliefScope：区分证据与用户压力对模型信念影响的受控框架

    用 BeliefScope 分离固定命题上证据与压力对模型信念的影响

    arXiv
    2610.11305
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Llama3.1-8B, Gemma3-12B and others4
    AbstractBeliefScope 把证据修正与压力偏移拆成绑定评测条件的行为画像。

    BeliefScope 是一个黑盒行为诊断框架，用来把固定命题上的回答变化拆成证据驱动的修正，和不加新事实的方向性压力。信念修正评测能看到信息基础变化后的更新，谄媚评测能看到用户影响下的顺从或漂移。作者认为，当两种输入都偏向同一结论时，表面偏移本身不能区分来源。

    Paper analysisRead analysis
  12. Evaluation and benchmarksarXiv 2610.11112

    EpiReal-Bench：商用图像生成模型虚假声明红队基准

    构建 EpiReal-Bench，评测图像生成器将虚假主张渲染为可信证据

    arXiv
    2610.11112
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GPT-Image-2, Nano Banana 2, Grok Imagen 2.0 and others4
    Abstract四款商用图像模型能把假主张画成可信证据。

    EpiReal-Bench 用来量商用图像生成器会不会把虚假现实主张画成可信视觉证据。

    Paper analysisRead analysis
  13. Evaluation and benchmarksarXiv 2610.11005

    研究：叙事包装可跨语言绕过 LLM 拒答

    提出 GUISE 基准与 AXIS 训练，防御叙事包装越狱

    arXiv
    2610.11005
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-1.7B, Qwen3-4B, GLM-4-9B
    AbstractGUISE 分开测包装与语域。

    作者测量叙事包装下的拒答失效，并训练一个在未见包装上拒绝、同时仍回答同风格良性请求的适配器。问题是：安全对齐模型拒绝直说的有害请求，却回答嵌进角色或叙事里的同一请求。文言文被用来把语域和包装拆开。

    Paper analysisRead analysis
  16. Evaluation and benchmarksarXiv 2610.10276

    PatchBench 提出激活补丁局部评测协议，揭示越狱修复的附带损伤

    提出 PatchBench-Local 协议，评测激活补丁修复越狱的局部附带损伤

    arXiv
    2610.10276
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2.5-3B-Instruct, Qwen2.5-14B-Instruct, gemma-3-4b-it and others8

    Abstract论文提出 PatchBench 与局部评估协议，揭示了激活补丁在全局能力无损下仍存局部良性退化。

    Paper analysisRead analysis
  17. Evaluation and benchmarksarXiv 2610.10232

    研究：九种 LLM 说服力评测方法排名仅弱相关

    适配多种已发表说服评测并比较模型排名

    arXiv
    2610.10232
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Claude-Fable-5.1, GPT-5.6-Sol, Kimi-K2.6 and others15
    Abstract说服分数混合能力与意愿，换任务后同一模型的排名可以差很多。

    这项研究比较的是自动化说服评测是否一致，而不是哪一种分数更正确。

    Paper analysisRead analysis
  18. 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
  19. Evaluation and benchmarksarXiv 2610.09964

    研究比较连续训练阶段对大语言模型说服力的影响

    比较连续后训练阶段如何改变语言模型的说服力

    arXiv
    2610.09964
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated, GPT-4
    Abstract作者称说服性 SFT 提高失调模型的态度变化，IPO 无额外收益，GPT-4 不异于中性文本。

    作者用同一人类协议，把 LLM 说服力拆成阴谋数据上的失调 SFT、说服数据上的 SFT，以及其后的 IPO。

    Paper analysisRead analysis
  20. 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
Load more