PR: Integrate DeepReviewer as a new arXiv Labs project #845
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
- Added DeepReviewer toggle to labs_tabs.html in the Demos section - Created deepreviewer.js to handle API communication and content display - Updated toggle-labs.js to support the new DeepReviewer component - Implemented background image and logo using base64 encoding - Styled UI element to match existing lab integrations - Added "AI Review" functionality similar to TXYZ integration
bdc34
approved these changes
Aug 12, 2025
Contributor
bdc34
left a comment
bdc34
left a comment
There was a problem hiding this comment.
The technical details of this labs application are acceptable.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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Outdated suggestions cannot be applied.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.
This PR proposes integrating DeepReviewer as an arXiv Labs project. I understand this may need further discussion, and I'm open to feedback or modifications if required.
DeepReviewer originated from the ICLR 2025 accepted paper "CycleResearcher: Improving Automated Research via Automated Review" and was recently enhanced with our improved version "DeepReview: Improving LLM-based Paper Review with Human-like Deep Thinking Process." The system employs a deep thinking process to deliver highly accurate and detailed analysis of scientific papers.
DeepReviewer has already demonstrated its value in the academic community through a collaboration with World Models @ ICLR 2025, where it helped enhance the usefulness of human expert reviews. Both our training data and models are fully open-sourced on Hugging Face, ensuring complete reproducibility and transparency.
This PR adds DeepReviewer to the AI Review tab in the arXiv Labs Panel, allowing readers to get instant, AI-powered analysis of papers directly through the arXiv interface. The integration works by transforming the current arXiv URL to our service URL (ai-researcher.net), which then provides detailed insights about the paper.
The key benefits for the arXiv community include:
This integration aims to make scholarly literature more accessible and understandable, helping researchers quickly grasp complex papers and decide which ones deserve deeper reading. All processing is done with full respect for the original content and proper attribution to the authors.
Here's what the DeepReviewer integration looks like on arXiv:
