Skip to content

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

Oct 13 · Tuesday

All papers

Found 51 papers
Filter4

Attack type

Jailbreaks18Prompt injection4Model tampering7Extraction and theft3Detection evasion9

Attack technique

Gradient and representation optimization48Automated search65Multi-turn escalation24Encoding and obfuscation52Role-play and persuasion27Conversation structure manipulation11Reasoning-chain manipulation11Cross-modal payloads47Side channels3Defense-targeted31Dormant triggers30Selection manipulation3

Impact

Harmful content24Data and privacy leaks3Output manipulation24

System type

Models and APIs51LLM applications9AI Agent14coding agent0web agent4Embodied systems and robotics6

Attacker access

Black-box12Gray-box3White-box19

Attack surface / entry point

User input19Images, audio and video23Weights and gradients10Inference services and hardware5

Layer

Prompt layer34Model layer11Infrastructure layer5

System component

Guardrails and judges3Vision18Audio3Multimodal22
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 203papers have no tags yet and are not included in filtered results.
Attack technique：Gradient and representation optimizationAttack technique：Side channelsSystem type：Models and APIsSystem type：coding agentClear all

Another 203papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.09027

    P-VMI：对抗图像经 KV cache 持续影响多轮对话

    提出持久视觉记忆注入，使对抗图像经缓存在触发时操控输出

    arXiv
    2610.09027
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    Qwen3-8B, Qwen3-VL-8B-Instruct, Gemma-4-12B

    Abstract论文提出 P-VMI，证实对抗输入被掩蔽后仍可经 KV 缓存持久控制模型，强调了保留状态的安全风险。

    Paper analysisRead analysis
  3. AttacksarXiv 2610.06848

    TranScope：利用 CPU 性能计数器对 Transformer 做成员推断攻击

    提出 TRANSCOPE，用 CPU 硬件计数器推断训练集成员

    arXiv
    2610.06848
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    BERT (4.4M-340M), ViT (22M-307M), Pythia (70M-2.8B) and others5

    AbstractTRANSCOPE 揭示了分词器在嵌入表引发的硬件访问局部性差异，证实微架构计数器可突破恒定时间与置信度遮蔽防御。

    Paper analysisRead analysis
  4. AttacksarXiv 2610.05894

    研究者提出闭环偏见注入攻击

    提出闭环激活引导，在扩散语言模型去噪过程中注入定向偏见

    arXiv
    2610.05894
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    Gray-box
    Evaluated models
    LLaDA-8B-Instruct, Dream-v0-Instruct-7B, LLaDA-MoE-7B-A1B-Instruct

    Abstract通过去噪中间概率反馈动态调节残差流转向，证明扩散语言模型在推理服务栈中存在闭环偏见注入脆弱性。

    Paper analysisRead analysis
  6. AttacksarXiv 2610.03430

    JIL：通过对抗后缀操纵 LLM 请求调度优先级

    提出 JIL 对抗后缀，压低长度预测以抬高请求调度优先级

    arXiv
    2610.03430
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3-8B-Instruct, Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3 and others4
    AbstractJIL 用对抗后缀压低长度预测，在预测式 LLM 调度中提高自身优先级。

    JIL 是对预测式 LLM 调度的攻击：用对抗后缀压低长度预测，使请求在 SRPT 式队列里更早得到服务。

    Paper analysisRead analysis
  7. AttacksarXiv 2610.03166

    LiBRA：通过双向隐空间优化实现检测感知的图像水印去除

    提出 LiBRA，在潜空间双向优化以规避图像水印双侧检测

    arXiv
    2610.03166
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Stable Signature, YU1, HiDDeN and others4

    AbstractLiBRA 用双向潜空间优化规避双侧水印检测，并提高相对反转基线的保真。

    Paper analysisRead analysis
  8. AttacksarXiv 2610.01367

    研究：高质量数据筛选挡不住投毒

    提出 Bi-QSTO，优化能穿过质量筛选并在微调中削弱安全对齐的投毒样本

    arXiv
    2610.01367
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Llama2-7B-Chat, Llama3-8B-Instruct, Qwen2-7B-Instruct and others5
    Abstract质量筛选挡不住部分高分样本的安全削弱。

    Bi-QSTO 研究一个实际训练缺口：基于质量的数据选择会先于微调决定样本去留，而以往微调投毒多假设毒样本直接进入训练集。已对齐 LLM 的拒答会被少量有害或看似无害的微调样本削弱。作者用三个问题串起实验：选择能否排除毒样本。

    Paper analysisRead analysis
  9. AttacksarXiv 2610.01062

    Kernelized Activation Steering：将激活引导推广到再生核希尔伯特空间的新框架

    提出核化激活引导 KAS，在推理时自适应转向激活以诱导越狱

    arXiv
    2610.01062
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2.5-7B-Instruct, Gemma-2-9B-IT, Llama-3.1-8B-Instruct and others4
    AbstractKAS 用核梯度做局部自适应转向，并在线性核下恢复 DiM。

    Kernelized Activation Steering（KAS） 把推理时的激活转向写成核空间中的优化，使更新随当前激活相对源、目标参考集的位置而变化。

    Paper analysisRead analysis
  10. InterpretabilityarXiv 2609.24801

    研究用 XAI 归因分析 Prompt Guard 2，同义词替换可翻转其提示注入判定

    用词级归因指导同义词替换，翻转提示注入检测并验证越狱

    arXiv
    2609.24801
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Evaluated models
    Prompt Guard 2 (86M), Llama 3.1 8B

    Abstract作者称检测靠分散的词法线索。

    Paper analysisRead analysis
  11. AttacksarXiv 2609.02172

    BOSS：以广度优先后缀搜索改进 GCG 越狱优化

    提出广度优先后缀搜索 BOSS，改进 GCG 越狱优化

    arXiv
    2609.02172
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box
    Evaluated models
    Llama-2-7B-Chat, Yi-1.5-9B-Chat, Qwen2-7B-Instruct and others8
    AbstractBOSS 把 GCG 类优化从单条深搜索改为多条短轨迹加源侧诊断。

    BOSS 是插在 GCG 类离散 suffix 优化上的搜索框架：局部梯度更新不变，变的是固定预算花在深度还是广度。

    Paper analysisRead analysis
  13. AttacksarXiv 2608.17485

    KeyPooling: Measuring where LLM API relay paths break prompt cache isolation

    Proposing KeyPooling to measure where API relays break prompt cache isolation

    arXiv
    2608.17485
    Published
    Layer
    Infrastructure layer
    Setting
    Models and APIs
    Evaluated models
    grok-4-fast-non-reasoning, grok-4-1-fast-non-reasoning, grok-3 and others11
    AbstractShared upstream credentials place relay customers in the same cache domain; preserving identity or separate principals can separate them.

    KeyPooling measures where LLM API relays drop customer identity, causing prompt caches to no longer be isolated per downstream customer.

    Paper analysisRead analysis
  14. AttacksarXiv 2608.07430

    研究者提出 SN-Guided Diffusion，利用扩散大模型安全神经元实现离线越狱

    提出 SN-Guided Diffusion，利用安全神经元实现离线越狱

    arXiv
    2608.07430
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box, White-box
    Evaluated models
    Qwen2.5-7B-Instruct, LLaDA-8B-Instruct, Dream-Instruct-7B and others15
    Abstract论文揭示了扩散语言模型对自回归安全神经元的继承性，并提出纯离线黑盒越狱框架 SN-Guided Diffusion。

    论文系统评估了扩散大语言模型（DLLM）在安全对齐中的双重角色——既作为白盒攻击目标，又作为黑盒越狱生成的对抗工具。

    Paper analysisRead analysis
  16. AttacksarXiv 2607.23496

    Concept2Scenario：用 SAE 概念归因发现可迁移的越狱场景

    提出 Concept2Scenario，借助内部概念归因发现可迁移的越狱场景

    arXiv
    2607.23496
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    White-box, Black-box
    Evaluated models
    Qwen3.5-9B, Llama-3.1-8B, Ministral-3-8B and others6

    Abstract该研究从表征空间因果归因揭示场景削弱拒绝的机制，发现可复用脆弱场景先验，提升多模型黑盒越狱效能与探索速度。

    Paper analysisRead analysis
  17. 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
  19. AttacksarXiv 2605.21541

    研究者提出 FRA-Attack 频域正则化方法，提升对闭源 MLLM 的迁移攻击效果

    提出 FRA-Attack，用频域对齐与梯度正则做可迁移定向视觉攻击

    arXiv
    2605.21541
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GPT-5.2, GPT-5.4, Claude-Opus-4.6 and others15
    AbstractFRA-Attack 以频域对齐和梯度低通做可迁移定向攻击，闭源面板上作者称整体更强，Gemma 除外。

    FRA-Attack 是面向闭源 MLLM 的迁移式定向图像攻击。作者要解决的是：空间域局部对齐重复全局低频，代理梯度的高频成分又把更新拉向代理自身。

    Paper analysisRead analysis
Load more