First off, thank you for considering contributing! Your help is essential for keeping this list up-to-date and comprehensive. We aim to curate high-quality resources dedicated to Large Language Models (LLMs) for mathematics.
There are several ways you can contribute:
- Adding new resources: Found a relevant paper, model, dataset, tool, or benchmark we missed? Please add it!
- Correcting existing entries: Notice a broken link, incorrect title, missing date, or wrong information? Help us fix it!
- Improving organization: Have suggestions for better sectioning or categorization? Open an issue to discuss!
- Check for Duplicates: Before adding a resource, please use the search function (
Ctrl+ForCmd+F) to ensure it hasn't already been listed. - Ensure Relevance: Submissions should be directly related to the intersection of Large Language Models and mathematics (reasoning, problem-solving, theorem proving, foundational capabilities, etc.).
- Fork the Repository: Click the 'Fork' button at the top right of the main repository page.
- Clone your Fork: Clone your forked repository to your local machine.
git clone [https://github.com/YOUR_USERNAME/Awesome-Math-LLM.git](https://github.com/YOUR_USERNAME/Awesome-Math-LLM.git) cd Awesome-Math-LLM - Create a New Branch: Create a descriptive branch name for your changes.
git checkout -b add-resource-xyz
- Add Your Contribution:
- Find the most relevant subsection in the
README.mdfile. - Add your resource entry, strictly adhering to the Formatting Guidelines below.
- Ensure the list within the subsection remains sorted in reverse chronological order (newest entries first).
- Find the most relevant subsection in the
- Commit Your Changes:
(Use a concise commit message describing your addition.)
git add README.md git commit -m "Add: [Resource Title] to [Section Name]" - Push to Your Fork:
git push origin add-resource-xyz
- Open a Pull Request (PR): Go back to your fork on GitHub and click the 'New pull request' button. Provide a clear description of the resource you added and why it's relevant. If your PR addresses an existing issue, link it (e.g.,
Closes #123).
Consistency is key! Please format your entries exactly as follows:
- Format:
* **[Optional Prefix]:** "[Resource Title]" ([Primary Link]) ([Secondary Link]) - *(Month Year)*
- Components:
*: Start with a bullet point.**[Optional Prefix]:**: Use a bolded prefix only if it identifies a well-known named model, dataset, technique, or concept for added clarity (e.g.,**Llemma:**,**Chain-of-Thought:**,**MATH Benchmark:**). Omit this if the title alone is sufficient."[Resource Title]": The official, accurate title of the paper, report, blog post, dataset name, etc., enclosed in double quotes. This is required.([Primary Link]): The main canonical link (e.g.,[Paper](URL_to_arXiv_abstract),[Blog Post](URL),[Website](URL)), enclosed in parentheses. Prefer official abstract/landing pages over direct PDFs.([Secondary Link]): An optional second link for code repositories, datasets, models, etc. (e.g.,([GitHub](URL)),([HF Dataset](URL)),([HF Models](URL))), also enclosed in parentheses.- *(Month Year)*: The first publication or release date (Month spelled out), enclosed in italicized parentheses (e.g.,- *(October 2023)*,- *(May 2024)*). If the month is unknown, use*(Year)*.
- Example:
* **Llemma:** "Llemma: An Open Language Model For Mathematics" ([Paper](https://arxiv.org/abs/2310.10631)) ([HF Models](https://huggingface.co/EleutherAI/llemma_7b)) - *(October 2023)*
- Ordering: Place your entry in the correct subsection based on its content, maintaining reverse chronological order (newest first) based on the
*(Month Year)*.
Feel free to open an issue for:
- Suggesting new sections or categories.
- Discussing significant changes or restructuring.
- Reporting broken links or errors you can't fix yourself.
- Asking questions about contributions.
Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms. Please be respectful and constructive in all interactions. (We recommend creating a CODE_OF_CONDUCT.md file, possibly using the Contributor Covenant template: https://www.contributor-covenant.org/)
Thank you again for your contributions!