logo
today local_bar
verifier

1 paper across 1 session

Poster Session 5

1 paper
Friday, December 5, 2025 · 11:00 AM → 2:00 PM
Exhibit Hall C,D,E
Tapered Off-Policy REINFORCE - Stable and efficient reinforcement learning for large language models
#206 · Nicolas Le Roux, Marc Bellemare, Jonathan Lebensold, Arnaud Bergeron, Joshua Greaves, Alexandre Fréchette, Carolyne Pelletier, Eric Thibodeau-Laufer, Sándor Tóth, Sam Work

A simple general purpose off-policy REINFORCE method which outperforms PPO, DPO and STaR on recent benchmarks.