Submit to Awesome Loop Models

PR Submission Wizard

Step 1

Pick what you are submitting

Choose Paper or Blog first. Papers must fit the loop-model scope. Blogs stay in the flat Blogs section.

Step 2

Fill metadata once

Pick the taxonomy sections (primary first) and, for papers, the loop topology, sharing, depth control, claims and comparison. Every value comes from a closed list; then review the generated path and YAML.

Step 3

Fork on GitHub and finish the PR

Copy the generated path and YAML, fork the repository on GitHub, create a branch in your fork, add the file at that path, then open a pull request back to this repo.

Loading the taxonomy and existing filenames from submission metadata…

Generate YAML and prepare your fork / PR

The form below generates the final YAML file path and content. GitHub does not provide a public no-auth link that can automatically fork this repo and create the file in your account, so the final step is manual: fork the repo, create a branch in your fork, create the file at the generated path, paste the YAML, and open a pull request.

Paper submissions need taxonomy sections and design axes. Blog submissions stay in the flat Blogs section, list taxonomy sections only, and use filenames of the form blogs/YYYY-MM-DD-shortname.yaml.

Paper Title Used in the YAML title field and for shortname fallback if needed. Authors The YAML generator will normalize this into an authors list. published_date

Use the calendar button or type plain YYYY-MM-DD. For blogs, the same date becomes the filename prefix. Venue / Source shortname (optional) If omitted, the page will derive a slug from the title. For non-arXiv papers, this becomes the filename. arXiv / Paper Link Code Link (optional) OpenReview Link (optional) Project Page (optional) One-line Description This becomes the YAML desc field.

Topic-map placement

Every entry lists the taxonomy sections it belongs to, primary first. The browser files the entry under its primary section; the other sections make it findable from their filters. GitHub pull requests do not support structured form fields, so these pickers stay local before handing you off to the manual fork / branch / PR flow.

Primary section

Required

The one section that best describes the entry's main contribution.

Other sections

0 selected

Optional. List them in order of relevance; the YAML keeps the order you pick them in.

Design axes

Describe where the loop sits, how weights are shared, how depth is controlled, what the paper claims looping buys and how it compares looped with untied models. Choose unspecified when the paper does not say.

survey_core is generated as false; maintainers set it for core loop-model references. must_read and catalog_fit are maintainer-only and omitted.

Taxonomy sections and design axes are closed vocabularies. If this submission needs a new value, describe the proposal in the PR message instead of inventing YAML values here.

Generated target path

Extracted arXiv ID Effective shortname Path status

Duplicate detection runs against the current YAML inventory from submission metadata.

Generated YAML

# Fill the required fields above to generate YAML.

GitHub fork / PR handoff

# Upstream GitHub repo URL will appear here.

Fork this upstream repo first. Then create the generated file path inside your own fork and paste the YAML shown above.

Manual GitHub step

Copy the generated target path and generated YAML from this page.
Open the upstream GitHub repo and fork it to your own account.
In your fork, create a branch and add a new file at the generated path.
Paste the YAML, commit it to your branch, then open a pull request back to this repo. If you need a taxonomy change, describe the proposal in the PR message.

JavaScript required for YAML generation

Without JavaScript, use TAGS.md and TAXONOMY.md manually and prepare the YAML by hand. Then fork the repo on GitHub, create a branch in your fork, create the file under papers/ or blogs/, paste the YAML, and open a pull request.