TIGER-AI-Lab/Program-of-Thoughts

Data and Code for Program of Thoughts [TMLR 2023]

40
/ 100
Emerging

This project helps AI researchers and developers improve how large language models (LLMs) solve numerical reasoning tasks. By giving the LLM a framework to express its thinking as a Python program, an external interpreter handles the actual calculations. This process takes a numerical reasoning problem as input and outputs a more accurate solution than standard prompting methods.

307 stars. No commits in the last 6 months.

Use this if you are developing or evaluating LLMs for tasks that require precise numerical computation and want to enhance their accuracy beyond typical Chain-of-Thought approaches.

Not ideal if you are a general user looking for a ready-to-use application, as this requires technical setup and an understanding of LLM development workflows.

AI research natural language processing numerical reasoning large language models prompt engineering
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

307

Forks

27

Language

Python

License

MIT

Last pushed

May 15, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/TIGER-AI-Lab/Program-of-Thoughts"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.