Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities

Pengyi Li, Elizaveta Goncharova, Andrey Kuznetsov +1

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an indispensable paradigm for enhancing reasoning in Large Language Models (LLMs). However, standard policy opt…