huggingface / huggingface/paperswithcode-feedback

Contribution workflow for adding benchmarks / tasks

Open
#1 18 comments 5 reactions 0 assignees View on GitHub
Dominant language
No language data
Stars
10
Forks
1
PR merge metrics
No merged PRs in 30d

Description

## How can the community contribute new tasks, benchmarks, and SOTA results?

Hi! Thanks for bringing the project back — really happy to see it alive again.

My HF profile: https://huggingface.co/DenisKochetov

I wanted to open a broader issue about contributions, not only for myself but for the whole research / open-source community.

The original Papers with Code was extremely useful because it covered many tasks, benchmarks, papers, and SOTA tables. It was also very convenient that many arXiv papers had direct Papers with Code links.

Maintaining benchmark coverage across many areas is a huge amount of work, so this feels like a perfect case for community contributions. Researchers are naturally motivated to add their own papers and SOTA results, especially when they want their work to be visible and properly listed.

So my main question is:

**What is the best way for users / researchers to contribute new tasks, benchmarks, datasets, papers, and results to the site?**

For example:

- adding a new task
- adding or editing benchmark datasets
- submitting a paper/result to a leaderboard
- adding a new leaderboard for an existing task
- fixing metrics, links, or reported numbers
- helping review community submissions

I’m personally interested in helping with 3D generation, image-to-3D, text-to-3D, 3D reconstruction, and animal pose estimation.

Would love to understand what contribution workflow you prefer: GitHub issues, PRs, forms, structured files, direct website submissions, or something else.

Happy to help.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.