ReST meets ReACT: improving ReAct with Self-Critique, AI Feedback, and Synthetic Data Generation.

A brief description of this adaptation of Reinforced Self-Training (ReST) to an agentic configuration.

This article describes a very interesting and inspiring proposal to improve a ReAct agent with reasoning and action response with external knowledge. Besides offering promising results, it seems to me that it presents workflows that can be used in many approaches, and therefore, I find it very useful to read and understand.

Introduction

In December 2023, Google researchers published the …

agent article artificial intelligence critique data feedback knowledge llm machine learning nlp react react agent reasoning rest self-training synthetic synthetic data training