Ai-Review

NeuroDong/Ai-Review is ranked 4,205 among these 8 peers; lukaszliniewicz/Pandrator is ranked 4,203.

Chinese 📢 Latest News **[20/08/2026]** Added **Relative Rank**: compare your manuscript with papers from a target venue (or ones you upload) and see where it stands. Evaluation on 100 pools: Examples/Relative_Rank_evaluation. 📖 About Ai-Review This repository is dedicated to using AI to optimize papers, making it easier for researchers to check the strengths, weaknesses, and improvement suggestions of manuscripts, and to gauge how competitive a paper is relative to comparable work. The following figure is a

usage preview: 🚀 Usage Tutorial 🌐 I. Basic review (LLM) **Entry**: **Steps**: Upload your PDF manuscript, choose a prompting mode (e. SoT, Few-Shot), and click to get structured Strengths, Weaknesses, Suggestions, etc. **Note**: You can use the quick demo on the page or configure your own API (e. VLM review **Entry**: **Steps**: Upload a PDF; the system captures page snapshots and runs VLM review, so layout and figures are taken into account. Best when you need the model to see figures

and equation layout. Prompt-injection detection **Entry**: **Steps**: Upload a PDF; the system checks whether the file contains hidden instructions meant to steer or override an AI reviewer. **Note**: Use this before relying on an AI review if you worry the PDF may have been tampered with. Relative Rank (competitiveness ranking) **Entry**: **Steps**: Upload your PDF manuscript. ICLR, NeurIPS) so the system can select comparison papers from that venue’s accepted list, or upload comparison papers yourself. Every PDF is

anonymized, then the pool is ranked by competitiveness so you can see where your manuscript stands. **Note**: This is a relative ranking among the papers in the pool, not a conference decision or an accept/reject prediction. Review tells you how to improve; Relative Rank tells you where you stand against comparable work. Evaluation on 100 pools (pairwise accuracy 83. 8%): see Examples/Relative_Rank_evaluation. Skills On platforms that support Agent Skills (e. Cursor), you can use Ai-Review

as a Skill and get reviews directly in the chat. Get and install The repo includes the Skill definition at: ai-review-skills/SKILL. md In Cursor: place the folder under in your project or in your user directory. Follow your platform’s instructions for loading Skills. Supported formats **LaTeX**: **PDF**: **Word**: / 3. How to use and trigger Say: use @ai-review-skills to review @paper. The agent will call the Ai-Review Skill to generate the review. 🔄 Continuous Update

This repository will be updated continuously. Welcome to use this repository, raise issues, and submit pull requests to help improve the prompt templates and website features, thereby helping the community improve paper quality and acceptance probability. 📄 Review Example Please see: Review example of "Deep Residual Learning for Image Recognition". Relative Rank evaluation (100 pools on PeerRead ICLR 2017): Examples/Relative_Rank_evaluation. 🧠 Prompt Engineering ***Reverse Prompt (Included)***: Infer prompts from excellent results generated by large models. The excellent results here

come from the AI review at AAAI2026. ***Few-Shot Prompting (Included)***: In the web UI, select "Prompt + Examples (Few-Shot