Group-relative REINFORCE Families #292
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
Summary of ChangesHello @yaochaorui, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly expands the reinforcement learning capabilities by integrating several "Group-relative REINFORCE Families." It introduces new algorithms focusing on clipping, importance sampling, regularization, and data-distribution shaping, along with baseline implementations. The changes include new advantage and policy loss functions, all designed with modularity in mind, and are accompanied by practical examples for common datasets like GSM8k and MATH. Highlights
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
Contributor
There was a problem hiding this comment.
Code Review
This pull request introduces implementations for several group-relative REINFORCE families (REC, REP, RED) and their usage examples. The changes include new algorithm types, advantage functions, and policy loss functions, along with example configurations for GSM8k and MATH datasets. The code is well-structured with a clear separation of concerns. My review identifies a couple of minor issues in the advantage calculation logic and a suggestion to reduce documentation duplication for better maintainability.
Collaborator
|
/unittest-module-common |
Summary
Tests
Github Test Reporter by CTRF 💚 |
pan-x-c
reviewed
Sep 22, 2025
pan-x-c
reviewed
Sep 22, 2025
1369c5c to
3950c30
Compare
Removed mention of multiple config files in README.
Removed reference to 'train_math.yaml' in README.
Collaborator
|
/unittest-module-common |
Summary
Failed Tests
Tests
Github Test Reporter by CTRF 💚 |
Collaborator
|
/unittest-module-common |
Summary
Failed Tests
Tests
Github Test Reporter by CTRF 💚 |
Collaborator
|
/unittest-module-common |
Summary
Tests
Github Test Reporter by CTRF 💚 |
Collaborator
|
/unittest-module-algorithm |
Summary
Tests
Github Test Reporter by CTRF 💚 |
pan-x-c
approved these changes
Sep 25, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
It includes three major families:
We also provide baseline implementations such as Vanilla REINFORCE and GRPO.
Checklist
Please check the following items before code is ready to be reviewed.