tml-epfl llm-adaptive-attacks issues - Githubissues

tml-epfl / llm-adaptive-attacks

Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks [arXiv, Apr 2024]

https://arxiv.org/abs/2404.02151

MIT License

224 stars 23 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Jailbreak artifacts cannot be reproduce the attack effect

#9 ZHIXINXIE opened 3 weeks ago
0
Questions about adversarial suffix generation

#8 Syyabb closed 3 months ago
1
Potential BUG

#7 Junjie-Chu closed 3 months ago
1
get_universal_manual_prompt template

#6 wusuhuang closed 4 months ago
2
Question about the tokenizer's pad_token when using llama2 as target model

#5 Kris-Lcq closed 4 months ago
2
Reproducing the experimental results

#4 bxiong1 closed 4 months ago
10
A typo in main.py

#3 franciscoliu closed 7 months ago
1
Question about the system prompt used for llama-2

#2 rickyang1114 closed 7 months ago
2
How to obtain the adv_init?

#1 xszheng2020 closed 7 months ago
2