This repository is dedicated to using AI to optimize paper, making it easier for researchers to check the strengths, weaknesses, and improvement suggestions of manuscripts.
Click here for a quick usage on web page. The following figure is a usage preview.
This repository will be updated continuously for a long time. Welcome to use this repository, raise issues, and push requests to help optimize this template to help better community members' papers get accepted.
Please see: Review example of "Deep Residual Learning for Image Recognition".
- Reverse Prompt Engineering (Included): Infer prompts through the outstanding results generated by large models. The excellent results here come from the AI Review of AAAI 2026.
- Few-Shot Prompting (Included): In the web UI, choose "Prompt + Examples (Few-Shot)" under "Prompting mode". The app will load templates from
Prompts/and include example reviews fromExamples/review_in_Resnet.mdandExamples/review_in_Verified.mdto guide the model. See few-shot prompting for background. - Chain-of-Thought Prompting (Ready to add): Let the model explicitly perform step-by-step reasoning before answering questions to improve the accuracy and logic of answers to complex tasks.
Please see here. We welcome your feedback on this prompt and help us optimize it to serve the community better.
- [21/10/2025] Added Few-Shot Prompting function to the webpage, see here.
- [18/10/2025] Updated the review effect samples, see here. These examples are generated using Few-Shot Prompting engineering techniques.
- [06/10/2025] The web version adds "Quick Try" (no need for users to set up API), and also allows users to set up Deepseek API.
- [02/10/2025] Updated the function of using Ai Review on the web page.
- [20/09/2025] Added math symbol and formula checks to enforce vulnerabilities.
- [14/09/2025] Added a review example in "Deep Residual Learning for Image Recognition".
- [14/09/2025] Optimize Prompt so that Ai generates more detailed Strengths, Weaknesses, and Suggestions in the form of a secondary list.