avatar
文章
55
标签
61
分类
10
首页
归档
标签
分类
智慧树
友链
关于
DoraemonblogJailbreaking Multimodal Large Language Models via Shuffle Inconsistency
搜索
首页
归档
标签
分类
智慧树
友链
关于

Jailbreaking Multimodal Large Language Models via Shuffle Inconsistency

发表于2025-08-24|更新于2025-08-24|TAI
|总字数:9|阅读时长:1分钟
Jailbreaking Multimodal Large Language Models via Shuffle Inconsistency
https://www.doraemonqwq.cn/2025/08/24/2025-8-24-Diffusion/
作者
Doraemon
发布于
2025-08-24
更新于
2025-08-24
许可协议
CC BY-NC-SA 4.0
周记JalibreakingDiffusion Model
投喂作者
  • wechat
    wechat
  • alipay
    alipay
cover of previous post
上一篇
Agent Attack
...
cover of next post
下一篇
Visual Contextual Attack Jailbreaking MLLMs with Image-Driven Context Injection
...
评论
avatar
Doraemon
你若盛开，清风自来
文章
55
标签
61
分类
10
🛴前往小家...
公告
欢迎来到我的小窝☘️☘️☘️
主域名：
doraemonblogs.github.io 
备用域名：
doraemonqwq.cn 
目录
  1. 论文部分
    1. 【ACM MM 2025】Manipulating Multimodal Agents via Cross-Modal Prompt Injection
    2. 【NeurIPS 2023】On Evaluating Adversarial Robustness of Large Vision-Language Models
    3. 【ICCV 2025】Jailbreaking Multimodal Large Language Models via Shuffle Inconsistency
    4. 总结：
  2. 实验部分
©2024 - 2025 By Doraemon
I wish you to become your own sun, no need to rely on who's light.
搜索
数据加载中