Amazon· Bedrock / NovaCustom reward functions for multi-turn reinforcement learning with Amazon Nova Forge
Amazon Nova Forge introduces custom reward functions for multi-turn reinforcement learning (RFT) using Group Relative Policy Optimization (GRPO), allowing developers to define reward logic through Bring Your Own Orchestration (BYOO) or a serverless option.
From the source
Amazon Nova offers multiple customization approaches, with reinforcement fine-tuning (RFT) standing out because it can teach models the behaviors you want through iterative feedback.
aws.amazon.com
Visit source ↗aws.amazon.com
- Announced
- 14 Aug 2026 16:02 UTC
Other stories
Amazon · 17 AugNVIDIA Nemotron 3.5 Lightning now available in Amazon SageMaker JumpStartAmazon · 17 AugBuild OpenClaw agents that transact with Amazon Bedrock AgentCore paymentsAmazon · 14 AugBuilding agentic workflows with SageMaker AI and Bedrock AgentCoreAmazon · 13 AugMonitor on-premises and multi-cloud AI agents with AgentCore Observability