yule-cai/aaai-review-simulator-prompt — explained in plain English
Analysis updated 2026-05-18
Paste your paper draft into an AI chat tool and get a mock AAAI-style review with numeric ratings before you submit.
Get problems sorted into fatal issues, major issues, and minor issues so you know what to fix first.
Use the one week revision plan to plan your last days before a deadline.
Check whether your paper's title, abstract, and body actually agree with each other.
| yule-cai/aaai-review-simulator-prompt | foxemsx/riptide | lucidrains/chroma-pytorch | |
|---|---|---|---|
| Stars | 159 | 159 | 159 |
| Language | — | Go | Python |
| Last pushed | — | — | 2022-12-27 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | easy | hard |
| Complexity | 1/5 | 1/5 | 5/5 |
| Audience | researcher | general | researcher |
Figures from each repo's GitHub metadata at analysis time.
This project is not a piece of software you install and run. It is a written prompt, a long block of instructions you paste into an AI chat tool like ChatGPT or Claude, that asks the AI to pretend it is a panel of AAAI conference reviewers looking at your research paper before you submit it. AAAI is a well known computer science and AI conference, and getting a paper accepted there is competitive. The prompt asks the AI to take on four separate reviewer personas plus a senior meta-reviewer, each looking at a different part of the paper: whether the method is technically sound, whether the experiments are solid and reproducible, whether the idea is actually new compared to existing work, and whether the writing and structure follow conference rules. Each reviewer gives numeric ratings and a confidence score, in the same style real conference reviews use. The stated goal is not to make the paper sound better or to predict whether it will actually get accepted. It is to catch real weaknesses before submission: to sort problems into ones that would likely get the paper rejected outright, ones that would seriously hurt the score but could be fixed with more work, and small polish issues that can wait until later. The prompt also asks the AI to avoid inventing evidence, page numbers, or citations it cannot verify, and to say plainly when something in the paper is not clear enough to judge. There are different versions of the prompt: a shorter Chinese version for a quick self check, and a longer, more thorough Chinese version meant for a serious pre-submission review. The output includes a full mock review report, a summary table of scores, and a day by day one week revision plan for fixing the most urgent problems first. This is aimed at researchers and graduate students preparing an AI or computer science paper, not at general developers. It requires no installation and no code, only copying the prompt text into an AI chat interface. The project is unofficial and has no connection to the real AAAI conference or its actual review process. It is released under the MIT license.
A copy-paste AI prompt that simulates a strict AAAI conference review panel, so researchers can find weaknesses in their paper before submitting it.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.