Skip to content

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

Oct 13 · Tuesday

All papers

Found 17 papers
Filter4

Attack type

Prompt injection8Poisoning and backdoors8Detection evasion3

Defense type

Model hardening27Inference-time intervention4Detection and filtering17Instruction–data separation21Permissions and approval14Monitoring and auditing7

Impact

Unauthorized actions4Output manipulation10

System type

Models and APIs7LLM applications5AI Agent5

Attack surface / entry point

User input38Web pages and interfaces9Documents and messages10Tool responses16Tool and skill descriptions10Code repositories and dependencies2Retrieval corpus5Memory writes3Inter-agent messages3Images, audio and video14Training data7Weights and gradients3Distribution channels8

Layer

Application layer10Training and data layer7

System component

Guardrails and judges8Vision3Multimodal3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Defense type：Detection and filteringAttack surface / entry point：Web pages and interfacesAttack surface / entry point：Code repositories and dependenciesAttack surface / entry point：Training dataClear all

Another 215papers do not have category tags yet and are not included in filtered results.

  1. Analysis and theoryarXiv 2610.13051

    Space beats statistics, retraining beats ranking: auditing training labels under label-only backdoors

    Auditing training labels for label-only backdoors, comparing spaces and statistics, and judging by retraining

    arXiv
    2610.13051
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    ResNet-32
    AbstractLabel inspection must be done in a space that has not been fitted to these labels, and judged by retrained models rather than detection scores.

    The authors ask: can a defender without trusted data decide which training labels to keep, and how much clean data this destroys, and what the final model accuracy is.

    Paper analysisRead analysis
  3. DefensesarXiv 2609.34804

    物理认证联邦学习：为关键水务基础设施的协同异常检测提供安全保障

    为协同异常检测提出物理准入门，在聚合前拦截不合规训练批次

    arXiv
    2609.34804
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    deep reconstruction autoencoder
    Abstract物理准入为联邦检测增加数据约束，但合规证明不等于无投毒或训练来源证明。

    这是一项面向水基础设施联邦异常检测的物理准入防御研究。 作者针对隐藏训练遥测带来的投毒检查问题，结合不变量挖掘、聚合前过滤和零知识抽样证明。

    Paper analysisRead analysis
  4. AttacksarXiv 2609.27124

    Fed-ADR：用恶意编排服务器协同客户端梯度操纵攻击联邦学习

    提出 Fed-ADR，用编排服务器协同操纵梯度攻破联邦学习并检测恢复

    arXiv
    2609.27124
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Attacker
    Gray-box
    Evaluated models
    CNN（MNIST：2 Conv + 3 FC）, CNN（Fashion-MNIST：6 Conv + 1 FC）, ResNet18（CIFAR-10）
    AbstractFed-ADR 用 gray-box 编排攻击绕过多种聚合防御，再用 Hessian 一致性检测和窗口内恢复把准确率拉回。

    Fed-ADR 是一套面向联邦学习的攻击、检测与恢复框架：编排服务器改写合谋客户端的梯度，使参数服务器的现有防御把投毒更新当成良性，同时在发现后用历史梯度把全局模型原地修回来。

    Paper analysisRead analysis
  5. AttacksarXiv 2609.22510

    特拉维夫大学提出爆炸性提示词：条件触发的间接提示注入可绕过九款生产 Agent

    提出条件触发的爆炸提示词，并设计摄入期检测器 DeFuse

    arXiv
    2609.22510
    Published
    Layer
    Application layer
    Setting
    AI Agent
    Evaluated models
    Llama-3.1-8B-Instruct, Llama-3.1-70B-Instruct, Qwen2.5-32B-Instruct and others15
    Abstract论文揭示了条件触发使提示注入绕过防御并在触发时执行工具，提出轻量检测器并在摄入期有效防范。

    论文针对大语言模型智能体面临的间接提示注入威胁，研究了利用条件句式实现时间分离的爆炸提示词。

    Paper analysisRead analysis
  7. DefensesarXiv 2609.07063

    VERITAS：抗投毒的本地差分隐私图学习协议

    提出 VERITAS，用双边背书剪除本地隐私图学习中的投毒节点

    arXiv
    2609.07063
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    GCN, GraphSAGE, GAT
    AbstractVERITAS 用分级信任的双边背书剪除 LDP 图学习中的投毒节点，并恢复分类准确率。

    VERITAS 是面向本地差分隐私图学习的投毒防御协议，按 trust-but-verify 在保持 ϵ-LDP 的同时剪除伪造节点。

    Paper analysisRead analysis
  8. DefensesarXiv 2609.06346

    面向后门攻击下持续学习的鲁棒动态扩展：净化与选择性恢复

    提出持续学习下的净化与选择性恢复框架，抵御逐任务训练投毒后门

    arXiv
    2609.06346
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    ViT-B/16
    Abstract作者报告任务内净化与恢复可配合动态专家防御后门，但保证依赖分离条件。

    这是一项面向持续图像分类的后门防御研究：模型只使用当前任务数据，同时需要保留过去知识并避免学习恶意监督。

    Paper analysisRead analysis
  10. Evaluation and benchmarksarXiv 2609.02316

    Counter-GEO-Bench：现有护栏难挡 GEO 虚假信息

    构建 COUNTER-GEO-BENCH，评测生成式搜索中的 GEO 虚假信息防御

    arXiv
    2609.02316
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    Gemma-4-31B-IT, Qwen-3.5-35B-A3B, Llama-4-Scout-17B-16E and others8

    Abstract论文构建了针对 GEO 虚假信息的首个防御基准，揭示通用护栏的失效与反作用，并提出了有效的轻量对比检测基线。

    Paper analysisRead analysis
  11. Evaluation and benchmarksarXiv 2609.05535

    Mind2Web-Injection：面向网页 Agent 视觉提示注入护栏的证据对齐评测

    提出 Mind2Web-Injection 评测网页 Agent 视觉注入护栏

    arXiv
    2609.05535
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    GPT-5.6-luna, Qwen3-VL-32B, Qwen3-VL-8B and others6
    Abstract作者称检测、定位和反事实响应在六个 VLM 上可分开。

    Mind2Web-Injection 用来检查网页截图护栏的判决是否对齐到该用的画面证据。。

    Paper analysisRead analysis
  13. Evaluation and benchmarksarXiv 2608.16824

    GEO-Flag：检测与测量面向生成式引擎优化的网页内容

    提出 GEOFlagBench 与 IPT，检测并测量生成式引擎优化网页

    arXiv
    2608.16824
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    ModernBERT-base, Qwen3-0.6B, Gemini-3.5-flash and others11
    AbstractGEOFlagBench 与 IPT 用于检测 GEO 页，并估计真实检索中的占比与引用可核验性。

    GEO-Flag 把页面级 GEO 检测做成可评测任务，并在发布的检索结果上估计被判为 GEO 的页面占比。

    Paper analysisRead analysis
  14. DefensesarXiv 2608.06422

    CMU 研究：分片判断可避免 LLM 监督失效并抵御对抗利用

    提出分片监督，将评判标准分配给独立调用以抵御展示攻击

    arXiv
    2608.06422
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Attacker
    Black-box
    Evaluated models
    Claude Opus 4.8, Claude Sonnet 4.6, Claude Opus 4.6 and others4
    Abstract论文揭示决策负载是大模型监督的独立瓶颈，提出分片与辩论机制有效分离注意力与平衡证据，在多项高难度基准上实现稳健评判。

    本文系统研究了大语言模型在执行多标准监督与评判时的决策负载瓶颈及其安全风险。

    Paper analysisRead analysis
  15. AttacksarXiv 2602.05484

    Masked Mirror: Covert Prompt Injection Attacks Against LLM-Based Phishing Detection

    提出隐蔽提示注入攻击 LLM 钓鱼检测器，并给出防御 InjectDefuser

    arXiv
    2602.05484
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Evaluated models
    OpenAI GPT-5, xAI Grok 4 Fast (Non-Reasoning), Meta Llama 4 Maverick and others4
    Paper analysisRead analysis
  17. Evaluation and benchmarksarXiv 2510.01354

    WAInjectBench 系统评测网页 Agent 的提示注入检测方法

    提出 WAInjectBench 评测网页 Agent 的提示注入检测

    arXiv
    2510.01354
    Published
    Layer
    Application layer
    Setting
    web agent
    Evaluated models
    GPT-4o, Mistral-7B-Instruct-v0.1, Llama-Prompt-Guard-2-86M and others8
    AbstractWAInjectBench 比较网页智能体的提示注入检测：显式线索可被部分检出，隐蔽攻击大多漏检。

    WAInjectBench 是网页智能体提示注入检测的基准。作者要看现有检测器能否发现这类攻击：攻击者改网页，使智能体执行目标动作，而不是用户任务。作者称这是首个此类综合基准。

    Paper analysisRead analysis
You have reached the end