Multi Turn RL
topic on 1 show · 2 statements across 1 episodes
2 statements about Multi Turn RL, every show
Current multi-turn RL research ignores discarded thinking tokens, breaking the math
“The existing paper people are writing about multi-turn RL are not actually incorporating this, and it kind of, like, breaks all the math.”
Long-horizon agent RL requires intermediate turn-level verification
“You probably want some kind of intermediate verification where you're not just waiting for the final answer at the end, But you want something like turn-level reward, potentially, where you want to be able to ensure that the model is getting credit for the mov…”