HomeResearch LibrarySelf-Refine: Iterative Refinement with Self-Feedback
system architectureChapter 5NeurIPS 2023 · 2023

Self-Refine: Iterative Refinement with Self-Feedback

Aman Madaan (CMU), Niket Tandon (Allen AI)

Abstract

We introduce Self-Refine, an approach for improving initial outputs from LLMs through iterative feedback and refinement. The same LLM generates the output, provides feedback, and refines the output.

Eigenvector Insight — Zone III / PASF-PADE AnalysisNot part of the original paper
Eigenvector Research — Marco van Hurne
How this paper contributes to solving the Zone III problem (PASF-PADE)

Self-Refine provides a simple but effective pattern for improving agent output quality at inference time. For enterprise deployments, this iterative refinement loop can be applied to any output that has a verifiable quality criterion.

Why AI is not sufficient for Zone III without this

Zone III refers to high-complexity, high-risk, long-running agentic workflows — the class of enterprise AI deployments where a single failure can cascade across hundreds of steps. Standard AI models, trained to predict the next token, are not inherently designed for durable, governed, multi-step execution. This paper addresses one or more of the structural gaps that make Zone III deployments unsafe without explicit architectural intervention.

Key Contributions

  • →Iterative self-refinement without training
  • →Self-feedback generation
  • →Quality improvement across diverse tasks

Topics

self-improvementiterative refinementfeedback loopsoutput quality