adding a new recipe on post training cr2 for driving captioning

[jingyijin2]

Description

Brief description of the changes in this PR.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Code refactoring
• Performance improvement

Changes Made

• Added/updated documentation
• Added/updated examples
• Fixed bugs or issues
• Improved code quality
• Updated dependencies

Testing

• I have tested the changes locally
• Documentation builds successfully
• Pre-commit hooks pass
• Examples run without errors
• Links and references are valid

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Additional Notes

Any additional information that reviewers should know.

[Copilot]

Pull request overview

This PR adds comprehensive documentation for post-training the Cosmos Reason 2 model for autonomous vehicle (AV) video captioning and visual question answering (VQA), developed in collaboration between Uber and NVIDIA. The recipe demonstrates how to adapt Cosmos Reason 2 for domain-specific AV captioning through targeted supervised fine-tuning.

Key changes include:

• A detailed post-training recipe with benchmark definition, zero-shot evaluation, data curation, fine-tuning, and re-evaluation sections
• Supporting assets (video samples, result visualizations) demonstrating training data and evaluation metrics
• Integration into the documentation structure with updated navigation files

Reviewed changes

Copilot reviewed 13 out of 13 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
docs/recipes/post_training/reason2/video_caption_vqa/post_training.md	Main recipe documentation covering workflow, benchmarks, training configuration, and evaluation results
docs/recipes/post_training/reason2/video_caption_vqa/assets/*.mp4	Video assets demonstrating training samples and annotation examples
docs/recipes/post_training/reason2/video_caption_vqa/assets/*.png	Visualization charts for BLEU, MCQ-based VQA, and LingoQA evaluation results
docs/recipes/post_training/reason2/video_caption_vqa/assets/*.json	Sample annotation file showing structured ground truth format
docs/recipes/post_training/reason2/video_caption_vqa/SUMMARY.md	Navigation file linking to the post-training documentation
docs/recipes/post_training/SUMMARY.md	Updated table of contents adding the new Reason 2 recipe entry
docs/recipes/inference/reason2/intbot_showcase/inference.md	Typo fix in existing documentation (unrelated to main PR purpose)

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated 1 comment.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 15 out of 15 changed files in this pull request and generated 2 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.