Skip to content

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

Oct 12 · Monday

All papers

Found 6 papers
Filter1

Defense type

Model hardening11Inference-time intervention6Detection and filtering35Instruction–data separation6Permissions and approval6Monitoring and auditing17

System type

Models and APIs3

Attack surface / entry point

User input4

Layer

Model layer3Application layer3
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 305papers have no tags yet and are not included in filtered results.
Defense type：Inference-time interventionClear all

Another 305papers do not have category tags yet and are not included in filtered results.

  1. 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
  3. Evaluation and benchmarksarXiv 2609.03438

    CONFLICTGUI 基准与 CONFLICTGUARD 框架：让多模态 GUI Agent 学会在冲突指令下终止执行

    提出 ConflictGuard，引导 GUI Agent 在冲突指令下终止执行

    arXiv
    2609.03438
    Published
    Layer
    Application layer
    Setting
    Browser and computer use
    Evaluated models
    UI-Venus-1.5-8B, UI-TARS-1.5-7B, OS-Atlas-Base-7B and others10
    AbstractConflictGUI 暴露 GUI 智能体的执行偏向过度服从，ConflictGuard 用条件转向提高冲突终止率。

    ConflictGuard用推理时的可行性核验和条件激活转向，让 GUI 智能体在指令不可行时终止，而不是盲执行。

    Paper analysisRead analysis
  4. Evaluation and benchmarksarXiv 2608.28233

    REINS：用 SAE 特征同时抑制有害续写并增强拒答的引导方法

    提出 REINS，在 SAE 空间抑制有害延续并增强拒答

    arXiv
    2608.28233
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3.5-4B-Base, Qwen3.5-2B-Base, Gemma-3-1B-IT
    AbstractREINS 分角色抑制有害延续并增强拒答。

    REINS 是一种推理时 SAE 安全转向方法，并配有用于上下文伪装的基准 GUISE。

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2606.22673

    Agent MechSuits：面向多轮 CLI Agent 的机制子空间安全引导

    提出 Agent MechSuits，对多轮 CLI Agent 做步级检测与子空间转向

    arXiv
    2606.22673
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    LLaMA-3.1-8B, Qwen-2.5-7B, Gemma-2-9B
    Abstract白盒步级探针加稀疏转向可压低三模型恶意动作，同时提高生成崩溃率。

    Agent MechSuits 是面向多轮 CLI 智能体的白盒、步级表示防御，并配有可执行轨迹基准 MAS。

    Paper analysisRead analysis
  6. Evaluation and benchmarksarXiv 2606.18673

    理解并缓解真实世界 LLM 应用中的提示词泄露攻击

    提出软提示防御 AREA，重锚定注意力以缓解系统提示泄露

    arXiv
    2606.18673
    Published
    Layer
    Application layer
    Setting
    LLM applications
    Attacker
    Black-box
    Evaluated models
    Llama-2-7B-chat-hf, Llama-3.1-8B-Instruct, Mistral-7B-Instruct-v0.3 and others8
    Abstract作者测量商用提示词泄漏，提出 attention drift，并用 AREA 在可用性与抗泄漏间取折中。

    作者研究真实 LLM 应用中的 system prompt 泄漏，并给出不改权重的缓解方法 AREA。

    Paper analysisRead analysis
You have reached the end