refactor(graders): improve parameter validation and streaming support

[XiaoBoAI]

• Add threshold validation in common graders (correctness, harmfulness, etc.)
• Fix streaming response handling in text_to_image grader
• Preserve metadata in grader score returns
• Update template default pattern (None with or fallback)
• Update model name in examples from qwen3-max to qwen3-32b

OpenJudge Version

[The version of OpenJudge you are working on, e.g. import openjudge; print(openjudge.__version__)]

Description

[Please describe the background, purpose, changes made, and how to test this PR]

Checklist

Please check the following items before code is ready to be reviewed.

• Code has been formatted with pre-commit run --all-files command
• All tests are passing
• Docstrings are in Google style
• Related documentation has been updated (e.g. links, examples, etc.)
• Code is ready for review

[gemini-code-assist]

Summary of Changes

Hello @XiaoBoAI, 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 refactors the grading system by improving parameter validation, particularly for threshold values, and enhancing the handling of streaming responses in multimodal graders. It also standardizes metadata preservation, updates template default behaviors, and ensures consistency in documentation and examples. These changes contribute to a more robust, user-friendly, and maintainable grading framework.

Highlights

• Parameter Validation: Implemented robust validation for the 'threshold' parameter across common graders (Correctness, Hallucination, Harmfulness, Instruction Following, Relevance) to ensure it falls within the acceptable range of [1, 5].
• Streaming Response Handling: Enhanced the text_to_image grader to correctly process streaming responses, ensuring that scores and reasons are accurately extracted from partial or streamed outputs.
• Metadata Preservation: Modified grader score returns to preserve existing metadata while adding new information, such as the 'threshold' used for evaluation.
• Flexible Template Defaults: Updated the default pattern for prompt templates in common graders, allowing None to be passed and gracefully falling back to the default template if no custom template is provided.
• Documentation and Example Updates: Standardized terminology in documentation by renaming 'ground_truth' to 'reference_response' and updated example model names from 'qwen3-max' to 'qwen3-32b' for consistency.
• Rubric Validation Improvement: Refined the rubric validation logic to correctly handle adjacent score ranges, preventing false positives for overlaps.

🧠 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 introduces several valuable improvements across the graders. The addition of threshold validation in common graders like CorrectnessGrader and HarmfulnessGrader enhances robustness. The pattern of using template or DEFAULT_TEMPLATE is a clean way to handle default prompt templates. Preserving existing metadata when adding the threshold is a good fix. The switch from logger.error to logger.exception will provide more context for debugging. The fixes in the text_to_image grader for streaming support and in criteria_utils are also solid. Overall, this is a great set of changes that improves code quality and consistency.