Skip to content
The day in AI
Wed 31 May 2023
Wednesday 31 May 2023
←
→
No written reading for this day yet. Here is everything that was logged.
Everything from this day
Research
31 May 2023
OpenAI
Let's Verify Step by Step (PRM800K)
Step-level feedback beats outcome-only feedback for training verifiers; best reward model solves 78% of a MATH subset; 800K step labels released.