research-idea-explainer
Translate a rigorously investigated research idea into intuitive motivation, a concrete scenario, a two-case example, and a causal story a graduate student can retell. Use as part of Research Idea Discovery final delivery or for requests like 我没懂这个idea.
- License 许可
- MIT
- 最近更新
- 2026-10-13
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- research-idea-explainer
- description
- Translate a rigorously investigated research idea into intuitive motivation, a concrete scenario, a two-case example, and a causal story a graduate student can retell. Use as part of Research Idea Discovery final delivery or for requests like 我没懂这个idea.
03 SKILL.md 正文
Idea Explanation Layer — 导师先让学生听懂
Read ../research-idea-discovery/references/plain-language-idea-protocol.md and ../research-idea-discovery/references/scientific-story-and-contributions.md.
Output, in this order:
- One-sentence daily-language question (≤50 Chinese characters when practical).
- Concrete task sample: exactly what inputs, classes/labels, and output the old model uses. If no grounded case is available, mark it a hypothetical teaching example.
- Two contrast situations holding one superficial property fixed (e.g. equally confident) but varying hypothesized task evidence. State what is observed versus hypothesized; do not invent measured results.
- What a competent ordinary baseline does, why it may fail in the selected case, and the most dangerous simple alternative that might already solve it.
- Core difference in one sentence, followed by technical terms/possible method—not before.
- A 30-second spoken scientific story and a falsifying E0.
Forbidden: empty slogans ("confidence ≠ evidence") without operational definitions; naming a graph without an estimable edge definition; claiming baselines actually fail without evidence. If the user says 没懂, make a smaller toy example and walk through concrete inputs and prediction changes instead of repeating vocabulary.
04 同仓库更多 Skills
research-review-board
暂无简介
research-proposal-generator
暂无简介
research-problem-validation
Validate research problems before generating ideas.
research-problem-scout
Use at the beginning of scientific idea discovery to find real, important, falsifiable research failures from actual sources; reject weak motivations before architectures and novelty-hunting. Includes low-cost problem-evidence scouting and a hard Motivation-First Gate.
research-pilot-designer
Use when user requests falsification pilot experiment design hypothesis testing research planning. Works within the general Research Idea Discovery plugin and any user-selected field.
research-paper-anchor-scout
Critically read a small adversarial group of anchor papers to discover real, falsifiable research problems from failures, successful-method mechanisms, contradictions and deployment assumptions; reuse verified assets without blindly modifying the previous network.