Workflow:ThreeSR Awesome Inference Time Scaling Manual Paper Contribution
| Knowledge Sources | |
|---|---|
| Domains | Academic_Curation, Open_Source_Contribution |
| Last Updated | 2026-02-14 01:30 GMT |
Overview
End-to-end process for contributing a new paper to the curated inference-time scaling list via the GitHub fork-and-pull-request workflow.
Description
This workflow covers the complete contribution path for adding papers to the repository when a contributor does not have direct push access. It follows the standard open-source fork, edit, and pull request pattern. Contributors can either use the automated script (fetch_semantic_info.py) within their fork or manually format the paper entry following the repository's template. The maintainer then reviews and merges the contribution.
The output is a merged pull request that adds one or more properly formatted paper entries to the README.md paper list.
Usage
Execute this workflow when you want to contribute a paper to the repository but do not have direct write access, or when you prefer the review-based contribution model. This is also the appropriate workflow when the paper is not yet indexed in Semantic Scholar and must be formatted manually, or when you want to add multiple papers in a single contribution.
Execution Steps
Step 1: Fork the Repository
Create a personal fork of the Awesome-Inference-Time-Scaling repository on GitHub. This provides a copy where you have full write access to make changes without affecting the upstream repository.
Key considerations:
- Fork from the main repository at ThreeSR/Awesome-Inference-Time-Scaling
- Clone the fork locally to work with the automated script
- Keep your fork synchronized with upstream before starting
Step 2: Add Paper Entry
Add the paper to the README.md paper list. There are two methods:
Method A (Automated): Run the fetch script against your local clone to auto-fetch and insert the paper:
Pseudocode:
- Run the fetch script with the paper title as argument
- The script queries Semantic Scholar, formats the entry, and inserts it
Method B (Manual): If the paper is not in Semantic Scholar or you prefer manual formatting, add an entry directly to README.md following the established template format: title with arXiv link, PDF link, authors, date, publisher, and collapsible abstract.
Key considerations:
- Follow the exact entry template used by existing papers
- Place the entry in the correct chronological position (newest first)
- Include all required fields: title, arXiv links, authors, date, publisher, abstract
Step 3: Verify Entry Format
Review the newly added paper entry to ensure it conforms to the repository's formatting standards. Check that all fields are populated, links resolve correctly, the date format is YYYY-MM-DD, and the entry is positioned correctly in the chronological order.
Key considerations:
- The emoji marker at the start of the title line is required for parsing
- The collapsible abstract uses HTML details/summary tags
- Empty venue fields should default to "arXiv.org"
Step 4: Open Pull Request
Commit the changes and push to your fork, then open a pull request against the upstream repository. The PR description should identify which paper(s) were added to facilitate review by the maintainer.
Key considerations:
- Reference the paper title in the PR description
- The maintainer will verify the addition before merging
- Issues can also be opened to request paper additions without making the edit yourself
Step 5: Maintainer Review and Merge
The repository maintainer reviews the pull request for correctness and completeness, then merges it into the master branch. This makes the paper entry available to all visitors of the repository.
Key considerations:
- The maintainer may request formatting corrections
- Duplicate papers will be flagged during review
- Alternatively, contributors can open an issue and let the maintainer handle the addition