Synapse Open Synapse
AI

Adversarial Attacks on LLMs

Lil'Log · · 33 min read

The use of large language models in the real world has strongly accelerated by the launch of ChatGPT. We (including my team at OpenAI, shoutout to them) have invested a lot of effort to build default safe behavior into the model during the alignment process (e.g. via RLHF ).

Latest from Lil'Log