A 7-Turn Failure: Assigning Blame in GRPO/GSPO
Multi-turn agentic RL has one reward per trajectory. How to split it across turns: the formalism, the credit-assignment spectrum, and GSPO-token’s opening.
Multi-turn agentic RL has one reward per trajectory. How to split it across turns: the formalism, the credit-assignment spectrum, and GSPO-token’s opening.