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

Jailbreaks5

Attack technique

Gradient and representation optimization27Automated search9Multi-turn escalation5Encoding and obfuscation14Conversation structure manipulation0Cross-modal payloads50Defense-targeted0Dormant triggers3

Impact

Harmful content5

System type

Models and APIs5

Attacker access

Black-box3

Attack surface / entry point

User input5Images, audio and video5

Layer

Prompt layer5

System component

MCP and skills9Guardrails and judges23Monitors6Reasoning chains7Permissions and sandboxing3Fine-tuning and open weights9Vision5Image generation3Multimodal10
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 240papers have no tags yet and are not included in filtered results.
Attack technique：Multi-turn escalationAttack technique：Conversation structure manipulationAttack technique：Defense-targetedSystem component：VisionClear all

Another 240papers do not have category tags yet and are not included in filtered results.

  3. AttacksarXiv 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