Paper Self Review logo

Paper Self Review

Community
HughYau
paper-self-review

Adversarially review a paper before submission and predict exactly what reviewers will attack, using the five-category rejection checklist (contribution, writing clarity, result strength, experimental coverage, method soundness) plus a claim-to-evidence audit. Produces a findings report with a verdict per item, not vague impressions. Use whenever the user asks if a paper is ready, what reviewers will say, whether the contribution is enough, whether they will get rejected, or asks you to review someone else's submission — and run it proactively before any submission deadline, because the checklist routinely surfaces a missing ablation or an unsupported abstract claim while there is still time to fix it. Boundary: this predicts what gets attacked and produces the findings list; actually rewriting what it finds is paper-revision. 中文触发:自评审、review论文、审稿、会不会被拒、contribution够不够、投稿前检查、adversarial writing、帮我看看这篇论文。

Overview

PublisherHughYau
Repositorypengsida-learning-research-skills
Skill namepaper-self-review
Stars
52
Forks
4
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by HughYau on GitHub. Read the source before you install it.

Installation

Install the Paper Self Review AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/HughYau/pengsida-learning-research-skills/tree/main/paper-self-review
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/HughYau/pengsida-learning-research-skills.git /tmp/pengsida-learning-research-skills
mkdir -p .claude/skills
cp -r /tmp/pengsida-learning-research-skills/paper-self-review .claude/skills/paper-self-review
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Paper Self Review in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Paper Self Review on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Paper Self Review is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

论文自评审

目的:确定审稿人可能指出的潜在问题,在投稿前就把它们解决掉。

方法叫 Adversarial writing:自己 review 自己的论文, 考虑 reviewer 可能会问的所有问题,并一一解决。


一、论文被接收需要做到三点

  1. Contribution 足够——需要包含以下几点中的若干: Novel task / Novel pipeline / Novel pipeline module / Novel design choices / New experimental findings / New insights
  2. 实验效果比之前的方法好。
  3. Ablation studies 和 comparison experiments 充分。

二、五方面拒稿检查清单

逐条过一遍,就知道这篇论文是否该被拒。每一条都要给出"是/否 + 证据",不要泛泛而过。

1. 技术贡献是否足够(论文有没有给读者带来新知识)

  • 1.1 想解决的 failure cases 是不是很常见?
  • 1.2 提出的技术是不是已经被 well-explored 了,该技术带来的 performance improvement 是可预见的 / well-known 的?
  • 1.3 技术是不是比较 straightforward?

如果 1.1 或 1.2 命中,问题出在选题而不是写作——回到 research-ideation 检查是否落入了"四种情况"的前两种。这时候改写作救不回来。

2. 论文写作是否清楚

  • 2.1 Introduction 是否清楚描述论文贡献?
  • 2.2 Pipeline figure 是否清楚描述 pipeline 与技术贡献?
  • 2.3 是否缺少技术细节、不可复现?
  • 2.4 每个方法模块是否都写了 motivation?

修复方法见 paper-revision。

3. 实验效果是否足够好

  • 3.1 是否只比之前的方法好了一点点?
  • 3.2 虽然比之前的方法好,但效果本身是否仍然不够好 / 不够 impressive?

4. 实验测试是否充分

  • 4.1 是否缺少 ablation studies?
  • 4.2 是否缺少重要的 baselines?是否缺少重要的 evaluation metric?
  • 4.3 数据是否太简单,无法证明方法是否真的 work?

5. 方法设计是否合理

  • 5.1 实验的 setting 是否不实际?
  • 5.2 方法是否存在技术缺陷、看起来不合理?
  • 5.3 方法的鲁棒性:是否需要在每个场景上调超参?
  • 5.4 新的方法设计在带来 benefit 的同时,是否引入了更强的 limitation, 导致新方法的净收益为负?

5.4 的判据:一篇论文要提升一个 metric,并且不明显损害其他 metrics。


三、必查的两条硬要求

1. 所有 claim 必须成立且有实验 support。

论文中所有的 claim(特别是 abstract 和 introduction 里的 claim), 需要不犯错,也得有实验 support。不然一些 reviewer 会直接以此拒掉论文。

做法:把 abstract 和 introduction 里的每一句断言列出来, 逐条标注它由哪个 table / figure / section 支撑。找不到支撑的,要么补实验,要么改写。

2. Related work 里必须讨论和自己方法最相关的论文。

漏掉最相关的工作,一些 reviewer 会直接以此拒稿。


四、输出模板

评审产出用这个格式。每一条都要带位置和证据—— "introduction 写得不够清楚"对作者没有任何用,"§1 第三段引出 technical challenge 时 直接跳到了我们的方法,没有说明 recent method 2 为什么会存在"才有用。

markdown
## 自评审:<论文标题> <日期>

**总判**:<接收 / 边缘 / 会被拒> 主要风险:<一句话>

### 命中项(按严重度排序)
| # | 类别 | 位置 | 问题 | 证据 | 修复动作 | 需要多久 |
|---|---|---|---|---|---|---|
| 1 | 4.1 缺 ablation | §5.2 | core contribution B 没有单独的 ablation | 表 3 只有 A+B 和 baseline | 补 A-only 一行 | 2 天实验 |
| 2 | 2.4 缺 motivation | §3.3 第一段 | 模块 C 直接给了设计,没说为什么需要它 | | 补一句 problem-driven 开头 | 30 分钟 |

### 五方面逐条判定
1. 技术贡献是否足够:<是/否 + 依据> 1.1 ☐ 1.2 ☐ 1.3 ☐
2. 写作是否清楚:2.1 ☐ 2.2 ☐ 2.3 ☐ 2.4 ☐
3. 实验效果是否够好:3.1 ☐ 3.2 ☐
4. 实验测试是否充分:4.1 ☐ 4.2 ☐ 4.3 ☐
5. 方法设计是否合理:5.1 ☐ 5.2 ☐ 5.3 ☐ 5.4 ☐
(☐ = 未命中即通过;☒ = 命中,需在上表出现)

### Claim 审计
| Abstract/Intro 里的 claim | 支撑它的 table/figure/section | 成立? |
|---|---|---|
|  |  |  |

### 时间可行性
距截稿 <N> 天。上表中来得及修的:<#…>;来不及的:<#…,建议怎么处理>

最后一节很重要:在截稿前三天报出一条需要两周实验的问题,等于没报。 来不及修的项要给出降级方案(改写 claim、加进 limitation、把强断言改成受限断言), 而不是让作者干着急。

五、Review 别人论文时

用完全相同的五方面清单和同一张表,逐条过一遍就能得出结论。 区别只是不需要"时间可行性"一节。


六、把自评审做成论文的一部分

在论文的最后加一个自我评审的 question lists,按上面五方面分别提问题, 然后根据这些问题改论文。改完再过一遍,直到清单上没有命中项。

追求完美主义是保证论文质量的非常重要的方式。


七、超出"不被拒"之后

清单只保证不被拒。要让论文收获好 review,还有一层:

把论文做得漂亮、美观,让人第一印象觉得这篇论文很高级。 三个抓手:好看的 teaser figure / pipeline figure;好看的表格和结果图;整齐的排版。

要让工作有影响力(而不只是被接收):把算法推到更难的数据上、面向下游研究群体做 demo,见 research-experiment; 选题层面的影响力判据见 research-ideation。

Frequently asked questions

What does the Paper Self Review AI skill do?

Adversarially review a paper before submission and predict exactly what reviewers will attack, using the five-category rejection checklist (contribution, writing clarity, result strength, experimental coverage, method soundness) plus a claim-to-evidence audit. Produces a findings report with a verdict per item, not vague impressions. Use whenever the user asks if a paper is ready, what reviewers will say, whether the contribution is enough, whether they will get rejected, or asks you to review someone else's submission — and run it proactively before any submission deadline, because the che...

Why use Paper Self Review on TypingMind?

Because you install it once and use it with any model. Paper Self Review is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Paper Self Review in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HughYau/pengsida-learning-research-skills/tree/main/paper-self-review. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Paper Self Review?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Paper Self Review?

As many as you like. As long as a model supports skills, you can use Paper Self Review with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Paper Self Review AI skill free?

It is published on GitHub by HughYau. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇