Skip to content

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

Oct 13 · Tuesday

All papers

Found 1 paper
Filter4

Attack technique

Gradient and representation optimization4Encoding and obfuscation1Conversation structure manipulation0Cross-modal payloads7

Attack surface / entry point

Images, audio and video1

System component

Guardrails and judges5Vision14Audio1Image generation3Multimodal16
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 215papers have no tags yet and are not included in filtered results.
Attack technique：Encoding and obfuscationAttack technique：Conversation structure manipulationAttack surface / entry point：Images, audio and videoSystem component：AudioClear all

Another 215papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2504.01094

    研究者提出 Multi-AudioJail

    提出 MULTI-AUDIOJAIL，以多语多口音和声学扰动越狱音频模型

    arXiv
    2504.01094
    Published
    Layer
    Prompt layer
    Setting
    Models and APIs
    Attacker
    Black-box
    Evaluated models
    Qwen2-Audio, DiVA-llama-3-v0-8b, MERaLiON-AudioLLM-Whisper-SEA-LION and others5

    Abstract黑盒多语多口音音频加声学扰动可抬高 LALM 的 JSR，文本示范防御并非在每个模型上都下降。

    Paper analysisRead analysis
You have reached the end