· 24 min read
Correct, minimal, and all
Many tasks have more than one minimal answer, none containing another. A reward that scores each rollout on its own cannot tell a new answer from a redundant copy of one it already has.
· 24 min read
Many tasks have more than one minimal answer, none containing another. A reward that scores each rollout on its own cannot tell a new answer from a redundant copy of one it already has.