madaan/self-refine
LLMs can generate feedback on their work, use it to improve the output, and repeat this process iteratively.
Implements a three-stage prompting architecture with separate Generator, Critic, and Refiner components that work across diverse tasks including code readability, math reasoning (GSM-8k), sentiment transfer, and visual diagram generation via GPT-4V. Uses `prompt-lib` for LLM querying and provides task-specific prompt templates (Init, Feedback, Iterate) that can be customized per application domain. Includes built-in evaluation scripts and supports iterative refinement with configurable stopping criteria across six benchmark tasks.
785 stars. No commits in the last 6 months.
Stars
785
Forks
68
Language
Python
License
Apache-2.0
Category
Last pushed
Oct 04, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/madaan/self-refine"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
genlm/genlm-control
Controlled text generation with programmable constraints
Intelligent-CAT-Lab/AlphaTrans
Artifact repository for the paper "AlphaTrans: A Neuro-Symbolic Compositional Approach for...
PCI-ORG/PCI-Personnel
Policy Change Index for Personnel (PCI-Personnel)
hemangjoshi37a/o1-meta-prompt
This project aims to emulate some of the advanced reasoning capabilities seen in models like...
gokmengokhan/deo-llm-reframing
Replication materials: Testing Distance-Engagement Oscillation as a prompting framework for...