AI.info

Philippe Morel

Explore Philippe Morel on AI.info.

• DeCoRL: Decoupling Reasoning Chains via Parallel Sub-Step Generation and Cascaded Reinforcement for Interpretable and Scalable RLHF