alt Hacker NewsGRPO is policy gradient/PPO with your value function baseline monte carlo estimated using k rollouts. The only new thing is finding out it works well with binary rewards and LLM policies.
GRPO is policy gradient/PPO with your value function baseline monte carlo estimated using k rollouts. The only new thing is finding out it works well with binary rewards and LLM policies.
It is a huge improvement to PPO because you don’t need a separate critic model which cuts memory costs in half and stabilizes training.