Skip to content

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

Oct 13 · Tuesday

All papers

Found 12 papers
Filter6

Attack type

Jailbreaks12Prompt injection93Poisoning and backdoors24Extraction and theft3Detection evasion24Malicious use6

Defense type

Inference-time intervention4

Impact

Harmful content7

System type

Models and APIs6AI Agent5

Attacker access

Black-box5White-box3

Attack surface / entry point

User input76Documents and messages3Tool responses15Tool and skill descriptions3Code repositories and dependencies2Images, audio and video7Weights and gradients6

Layer

Prompt layer43Model layer6Application layer6

System component

Vision4Audio3Multimodal7
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 201papers have no tags yet and are not included in filtered results.
Attack type：JailbreaksAttack surface / entry point：Documents and messagesAttack surface / entry point：Code repositories and dependenciesAttack surface / entry point：Images, audio and videoLayer：Model layerLayer：Application layerClear all

Another 201papers do not have category tags yet and are not included in filtered results.

  1. DefensesarXiv 2610.09703

    研究揭示 Token-Pruning 对 VLM 安全性的影响并提出 SAP 缓解机制

    提出 Safety-Aware Pruning，缓解视觉 token 剪枝放大的多模态越狱

    arXiv
    2610.09703
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    LLaVA-1.5-7B
    Abstract剪枝去掉背景 token 后注意力塌向恶意前景。

    这篇工作评估视觉 token 剪枝如何改变 VLM 在多模态越狱下的行为，并给出推理时防御 SAP。

    Paper analysisRead analysis
  7. AttacksarXiv 2607.11698

    AHA 用自动研究循环发现生产 Agent 漏洞概念

    提出 AHA 自动科研循环，发现生产 Agent 的可复用漏洞概念

    arXiv
    2607.11698
    Published
    Layer
    Application layer
    Setting
    coding agent
    Attacker
    Black-box
    Evaluated models
    Minimax-M2.7, Kimi-K2.6, Deepseek-V4-Pro and others5
    Abstract论文提出基于可证伪科研循环的 AHA 框架，从生产级智能体中挖掘并沉淀漏洞概念图，提升攻击复用率并指导防御加固。

    AHA 是一项针对生产级 LLM 智能体自动化红队测试的研究，旨在解决传统方法仅保留具体攻击载荷而缺乏成因解释、导致跨版本复用困难且无法指导修复的问题。

    Paper analysisRead analysis
  9. Evaluation and benchmarksarXiv 2602.14161

    When benchmarks lie: evaluating malicious prompt classifiers under real-world distribution shift

    提出 LODO 评测恶意提示探针的跨数据集泛化

    arXiv
    2602.14161
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.1-8B-Instruct, Gemma-3-27B, Qwen-3.5-4B and others5
    Abstract作者称 LODO 露出 CV 高估，数据集身份与安全方向部分重叠，标准域泛化干预关不上。

    LODO 用来检查恶意提示词分类器换成没见过的数据集后，还剩多少检测能力。。

    Paper analysisRead analysis
  10. DefensesarXiv 2602.11096

    Safety Recovery of Reasoning Models Only Needs a Few Early Steps of Guidance

    Proposes SafeThink, an inference-time defense that monitors the chain of thought and injects a redirect prefix when it falls below a threshold.

    arXiv
    2602.11096
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    R1-Onevision-7B, OpenVLThinker-7B, VLAA-Thinker-7B and others6
    Paper analysisRead analysis
  12. Evaluation and benchmarksarXiv 2510.01359

    JAWS-Bench：研究者用三种工作区场景系统越狱代码 Agent

    提出 JAWS-Bench，评测代码 Agent 在三级工作区中的越狱与恶意代码可执行性

    arXiv
    2510.01359
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-4.1, GPT-o1, DeepSeek-R1 and others8

    Abstract论文提出了评估代码智能体端到端恶意代码越狱的 JAWS-Bench，发现多文件与智能体推理会明显放大可部署危害。

    Paper analysisRead analysis
You have reached the end