Skip to content

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

Oct 12 · Monday

All papers

Found 5 papers
Filter4

Attack type

Jailbreaks4

Attack technique

Gradient and representation optimization1Multi-turn escalation4Encoding and obfuscation9Role-play and persuasion5Cross-modal payloads3

Impact

Harmful content4

System type

Models and APIs5

Attack surface / entry point

User input4

Layer

Prompt layer5Model layer0Application layer4

System component

Guardrails and judges3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 232papers have no tags yet and are not included in filtered results.
Attack technique：Gradient and representation optimizationAttack technique：Multi-turn escalationLayer：Prompt layerLayer：Model layerClear all

Another 232papers do not have category tags yet and are not included in filtered results.

  1. Evaluation and benchmarksarXiv 2609.36849

    研究评估 GradSafe 在多轮对话中的越狱检测脆弱性

    评测多轮对话中 GradSafe 越狱检测的可分性与脆弱性

    arXiv
    2609.36849
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Llama 3.1 8B Instruct, Qwen2.5-7B-Instruct
    AbstractCWS 显示不安全梯度在真实多轮流量中变弱，且随攻击族和模型而变。

    作者评估梯度越狱检测从单条提示词迁到多轮对话后还剩多少可分性，方法是用 Context Window Scanner 把 GradSafe 原封不动地套到用户轮窗口上。

    Paper analysisRead analysis
  2. Evaluation and benchmarksarXiv 2607.19855

    研究者提出最小范数攻击集成，用鲁棒性曲线替代单一 ε 评测

    提出最小范数攻击集成，用鲁棒性曲线评测图像分类器

    arXiv
    2607.19855
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    C1–C20（CIFAR-10，Table 1）, I1–I10（ImageNet，Table 1）
    Abstract预算可控的最小范数集成用 AOI 逼近攻击前沿，DOI 给出不依赖单一ε的防御排序。

    这篇工作把对抗鲁棒评测从单一 ε、固定集成，改成可按查询预算收紧的曲线估计。

    Paper analysisRead analysis
  3. Evaluation and benchmarksarXiv 2606.02630

    MultiTurnPSB

    构建 MultiTurnPSB，评测四轮医疗越狱与分类器输入防御

    arXiv
    2606.02630
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    GPT-4.1-mini, Claude Sonnet 4.5
    Abstract多轮攻击加剧医疗模型风险，分类器干预有效但受高误报率制约。

    MultiTurnPSB 针对患者端医疗大模型在持续交互中的安全风险，构建了包含三种攻击模态的四轮对抗评测基准。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2603.10068

    ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models

    Using ADVERSA to measure guardrail degradation and reviewer reliability under multi-turn attacks

    arXiv
    2603.10068
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Claude Opus 4.6, Gemini 3.1 Pro, GPT-5.2
    AbstractADVERSA measures multi-turn guardrail changes using per-turn compliance trajectories and three-reviewer consensus.

    ADVERSA is an automated multi-turn red-teaming measurement framework that records guardrail behavior as per-turn compliance trajectories and simultaneously measures whether reviewers agree. The problem it addresses: single-turn binary pass/fail cannot show how safety properties change under sustained adversarial pressure, nor can it separate gradual refusal consolidation from stable scores. The authors argue that real-world adversaries rewrite and persist over multiple turns.

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2603.20198

    研究者提出 MM-Plan 多模态越狱框架

    提出 MM-Plan 多模态越狱框架，自动规划多轮图像编辑以越狱视觉独占目标

    arXiv
    2603.20198
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Llama-3.2-11B, InternVL3-8B, Qwen3-VL-8B and others8
    Abstract作者称全局规划能攻破对单轮攻击更稳的前沿 MLLM。

    MM-Plan 是面向 Visual Exclusivity 的 black-box 多模态红队方法，并配有人工基准 VE-Safety。

    Paper analysisRead analysis
You have reached the end