Quantum Many-Body Physics Calculations with Large Language Models

This repository hosts the code and data associated with the paper "Quantum Many-Body Physics Calculations with Large Language Models."

Installation and Execution

• Requirement: openai version 0.28.1.
• Configuration: Place your OpenAI API key in a .env file in the root directory, using the format OPENAI_API_KEY=your_key_here.
• Running the Code: Navigate to the directory named after the arXiv number of interest and execute python ../utils.py ../Prompt_template.md {arxiv_number}, replacing {arxiv_number} with the actual arXiv number.

File Structure

Each arXiv paper is represented by a directory containing:

• {arxiv_number}.pdf: The arXiv paper.
• {arxiv_number}.tex: LaTeX source for the paper.
• {arxiv_number}_SM.pdf: Supplemental material.
• {arxiv_number}_SM.tex: LaTeX source for the supplemental material.
• {arxiv_number}.yaml: Configuration file with placeholders, extraction, execution results, and scoring details.
• {arxiv_number}_auto.md: Execution output.
• {arxiv_number}_extractor.md: Extraction output.
• {arxiv_number}_extractor.ipynb: Extraction helper functions.
• {arxiv_number}_execution.ipynb: Execution helper functions.
• {arxiv_number}_score_prompt.ipynb: Scoring and correction helper functions.

Printouts

Located in printout:

• {arxiv_number}_execution.pdf: Human-readable execution result.
• {arxiv_number}_extraction.pdf: Human-readable extraction result.
• {arxiv_number}_execution.tex: LaTeX source for execution result.
• {arxiv_number}_extraction.tex: LaTeX source for extraction result.
• {arxiv_number}_execution.md: Markdown for execution result.
• {arxiv_number}_extraction.md: Markdown for extraction result.

Additional Resources

• Rubrics.md: Scoring rubrics.
• Naming.yaml: Mapping of the task names used in Prompt_template.md and in the paper for readability.
• processed_oneshot_df_reverified_scores_renamed.yaml: Results of one-shot extraction.
• Prompt_template.md: Prompt templates.
• Task_type.yaml: Task types.

Authors

Haining Pan, Nayantara Mudur, Will Taranto, Maria Tikhanovskaya, Subhashini Venugopalan, Yasaman Bahri, Michael P. Brenner, Eun-Ah Kim