Dark mode
Light mode
Tag: source/paper
7 items with this tag.
Feb 18, 2024
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
seed
source/paper
Feb 17, 2024
Are Aligned Neural Networks Adversarially Aligned
seed
source/paper
Feb 15, 2024
AmbigQA - Answering Ambiguous Open-domain Questions
seed
source/paper
Feb 15, 2024
We're Afraid Language Models Aren't Modeling Ambiguity
seed
source/paper
Feb 14, 2024
Jailbroken - How Does LLM Safety Training Fail?
seed
source/paper
Feb 14, 2024
Multi-step Jailbreaking Privacy Attacks on ChatGPT
seed
source/paper
Feb 14, 2024
Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks
seed
source/paper