Evan Ellis
@evandavidellis
Research @ Applied Compute | BAIR | Scale | Imbue
814 Following 318 Followers
We're adding reward-hacking monitors into AC2. Every RL run gets checked for cheating, and our research agent, Ari, automatically investigates the flags.
After a model discovers an exploit, it can become its default strategy within a few training steps. We used to hunt for reward hacking manually, reading traces and running one-off agent queries. This was error prone and didn't scale. We decided to automate detection instead.
Show more
