[algo] fix gspo sequence ratio broadcast

[rank-Yu]

fixs #595

[gemini-code-assist]

Summary of Changes

Hello @rank-Yu, 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 resolves a specific bug related to tensor broadcasting within the compute_policy_loss function, particularly affecting the calculation of the GSPO sequence ratio. By adjusting the dimensionality of a key tensor, the change ensures that subsequent element-wise operations are performed correctly, leading to accurate loss computation.

Highlights

• Fix GSPO Sequence Ratio Broadcasting: Addressed a broadcasting issue in the calculation of the GSPO sequence ratio within the compute_policy_loss function.
• Tensor Shape Correction: Applied a .unsqueeze(-1) operation to negative_approx_kl_in_seq to ensure correct tensor shape for element-wise multiplication with response_mask.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The 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 /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

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 .gemini/ folder in the base of the repository. Detailed instructions can be found here.

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

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request addresses a critical bug in the 'gspo' loss calculation within compute_policy_loss. The change correctly fixes a tensor broadcasting error by adding .unsqueeze(-1) to negative_approx_kl_in_seq. This ensures the sequence-level importance ratio can be correctly multiplied with the response_mask, preventing a runtime crash. The fix is accurate and essential.

[hiyouga]

LGTM