AWS introduces search agent fine-tuning with multi-turn RL on SageMaker AI
AWS announced fine-tuning of search agents with multi-turn reinforcement learning (MTRL) on Amazon SageMaker AI. The approach optimizes the agent across the full multi-turn trajectory rather than single responses, aiming to combine a small model's speed and cost with the reliability that would otherwise require a frontier model. AWS shared the results it observed in retrieval quality and reliabili
