Skip to content

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

Oct 13 · Tuesday

All papers

Found 16 papers
Filter3

Attack type

Jailbreaks9Poisoning and backdoors7

Attack technique

Gradient and representation optimization26Automated search47Multi-turn escalation22Encoding and obfuscation42Role-play and persuasion19Conversation structure manipulation9Reasoning-chain manipulation7Cross-modal payloads34Defense-targeted10Dormant triggers7

Impact

Harmful content16Unauthorized actions15Data and privacy leaks13Output manipulation29

System type

Models and APIs15

Attacker access

Black-box6

Attack surface / entry point

User input11Training data3Weights and gradients4Distribution channels3

Layer

Prompt layer8Training and data layer6

System component

Fine-tuning and open weights6
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 227papers have no tags yet and are not included in filtered results.
Attack technique：Conversation structure manipulationAttack technique：Dormant triggersImpact：Harmful contentClear all

Another 227papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.07723

    Researchers propose answer-side backdoors: letting the model self-generate trigger words to bypass safety refusals.

    提出回答端后门，让模型自生成触发词以在干净有害请求下绕过拒答

    arXiv
    2610.07723
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Llama-2-7B-chat-hf, Mistral-7B-Instruct-v0.3, Gemma-7B-it and others4

    Abstract论文提出了多轮对话中回答端自生成触发词绕过安全拒绝的后门攻击，指出输入端防护存在盲区。

    Paper analysisRead analysis
  2. AttacksarXiv 2610.07654

    研究揭示策略蒸馏的安全后门风险

    测量带后门教师经 OPD 向干净学生迁移潜伏后门

    arXiv
    2610.07654
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Qwen2.5-1.5B-Instruct, Qwen2.5-3B-Instruct, Qwen2.5-7B-Instruct
    Abstract作者称 OPD 可在压低无触发词有害率时迁入后门。

    带后门的外部教师经 OPD，可在压低无触发词有害率的同时，把触发条件下的有害行为传给初始干净学生。。

    Paper analysisRead analysis
  3. AttacksarXiv 2609.36956

    研究者提出 Controlled Decoding Attacks，通过纯文本接口越狱黑盒 LLM

    提出 BLINDBIAS，通过纯文本续写与前缀控制越狱黑盒模型

    arXiv
    2609.36956
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    GLM-5, Gemini-3.5-Flash, Qwen3-32B and others4
    Abstract作者认为，隐藏数值概率并不足以消除支持重复采样和前缀续写的解码攻击面。

    BLINDBIAS 是一种针对 sample-only 续写接口的 black-box 解码时越狱框架，研究没有目标数值概率时，采样能否支持逐步控制。

    Paper analysisRead analysis
  6. AttacksarXiv 2609.07048

    FreqDoor：面向视觉语言模型的频域后门攻击

    提出 FreqDoor，在频域混合振幅并对生成式 VLM 做训练时视觉后门

    arXiv
    2609.07048
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    BLIP-2 (opt-2.7b), InstructBLIP (flan-t5-xl), LLaVA (interleave-qwen)
    AbstractFreqDoor 用频域振幅混合给三款 VLM 植入仅视觉的训练时后门。

    FreqDoor 是针对生成式 VLM 的仅视觉、训练时后门：在频域混合触发源振幅、保留干净图像相位，使触发在空间上分散。

    Paper analysisRead analysis
  9. Analysis and theoryarXiv 2606.04168

    论文提出随机插入攻击，利用自回归一致性绕过 LLM 安全对齐

    解释自回归一致性如何导致浅层安全对齐，并提出随机插入攻击

    arXiv
    2606.04168
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Abstract作者用自回归一致性串起浅层对齐、插入攻击和最坏延续对齐。

    作者把浅层安全对齐和一类续写攻击都归到自回归一致性上。

    Paper analysisRead analysis
  10. AttacksarXiv 2604.00627

    When Safe Models Become Dangerous: Exploiting Potential Vulnerabilities in LLM Merging

    提出 TrojanMerge，在源模型植入仅在合并后生效的潜伏分量

    arXiv
    2604.00627
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Llama 2 M1 (safety-aligned chat variant), Tulu-2-7b, Llama 3 M1 (safety-aligned chat variant) and others9
    Paper analysisRead analysis
  11. AttacksarXiv 2603.12681

    Colluding LoRA: A Compositional Vulnerability in Safety Alignment of Large Language Models

    提出 CoLoRA，让单独良性的适配器合并后抑制拒答

    arXiv
    2603.12681
    Published
    Layer
    Training and data layer
    Setting
    Models and APIs
    Evaluated models
    Gemma2-2B-It, Qwen2.5-7B-Instruct, Llama3-8B-Instruct
    Abstract作者用 CoLoRA 报告：单适配器可保持拒答，特定合并后标准有害提示上的遵从升高。

    这篇工作给出一个供应链场景下的组合失败：单独看起来可用、也会拒答的 LoRA，可以在用户按标准线性合并把它们同时装上之后，才对未修改的有害请求失去拒答。

    Paper analysisRead analysis
  12. InterpretabilityarXiv 2603.08234

    The Continuation vs. Refusal Debate: Mechanism Analysis of Continuation-Triggered Jailbreaks in Large Language Models

    Analyzing the attention head mechanism of continuation-triggered jailbreaks and proposing HCS

    arXiv
    2603.08234
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    LLaMA-2-7B-Chat, Qwen2.5-7B-Instruct
    AbstractThe authors state that jailbreaks arise from competition between two types of heads, and that HCS plus distillation can use this competition for inference-time intervention.

    The authors use attention head mechanism analysis for continuation-triggered jailbreaks: under the same harmful instruction, moving the continuation suffix from after the instruction to outside the prompt boundary makes the model more likely to continue generating along the suffix.

    Paper analysisRead analysis
  15. AttacksarXiv 2510.10271

    MetaBreak：通过特殊 token 操纵越狱在线 LLM 服务

    提出 MetaBreak，用特殊 token 操纵越狱在线 LLM 服务并绕过审核

    arXiv
    2510.10271
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Llama-3.3-70B-Instruct, Qwen-2.5-72B-Instruct, Gemma-2-27B-Instruct and others10

    Abstract利用特殊 token 注入对抗平台模板包裹与审核，多场景验证了越狱有效性与互补性。

    Paper analysisRead analysis
You have reached the end