Remove Redundant Assignment in Qwen3_VisionPatchMerger

[LJH-LBJ]

Purpose

This PR refactors the Qwen3_VisionPatchMerger class to remove redundant assignments related to use_postshuffle_norm and the initialization of the normalization layer.

Test Plan

Test Result

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors.

You ask your reviewers to trigger select CI tests on top of fastcheck CI.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

If you have any questions, please reach out to us on Slack at https://slack.vllm.ai.

🚀

[gemini-code-assist]

Code Review

This pull request aims to refactor Qwen3_VisionPatchMerger to remove redundant assignments. However, the current change introduces a critical bug by removing the assignment to self.use_postshuffle_norm, which is used in the forward method. This will lead to an AttributeError at runtime. Additionally, the logic for initializing the normalization layer is now incorrect when use_postshuffle_norm is true, which would cause a tensor shape mismatch. I've provided a suggestion to fix this.

[gemini-code-assist]

This change introduces a bug. By removing the assignment to self.use_postshuffle_norm, the forward method will raise an AttributeError because it relies on this attribute.

Furthermore, the normalization layer self.norm is now always initialized with context_dim. However, when use_postshuffle_norm is True, it should be initialized with self.hidden_size to match the tensor shape in the forward method. The current implementation will cause a shape mismatch error.

To fix this, you should restore the assignment and the conditional logic for the normalization dimension. Here is a suggested correction that keeps the logic correct while improving readability slightly.

        self.use_postshuffle_norm = use_postshuffle_norm
        norm_dim = self.hidden_size if use_postshuffle_norm else context_dim
        self.norm = norm_layer(norm_dim)

[LJH-LBJ]

self.use_postshuffle_norm is initialized in L220 and context_dim will be equal to self.hidden_size if self.use_postshuffle_norm is true in L221

[ywang96]

Ah thanks for finding this out!

[Isotr0py]

Thanks!