AI Paper Highlights | 论文要点速览

Turn a research paper (arXiv ID, title, or URL) into a tight bilingual summary of the things that matter: what it proposes, how it works, what it shows, and what it does not solve.

把一篇论文（arXiv ID / 标题 / 链接）提炼成简短的双语要点：提出了什么、怎么做的、结果如何、局限在哪。

When to use | 适用场景

• The user gives a paper link or arXiv ID and asks for a summary / 要点
• The user asks 「这篇论文讲了什么」「有什么最新研究值得关注」
• Comparing or shortlisting papers for deeper reading

Workflow | 工作流程

1. Identify the paper — extract the arXiv ID (e.g. 2401.00001) or resolve the title.
2. Fetch the abstract — WebFetch https://arxiv.org/abs/<id> (or the provided URL). If the URL is not arXiv, fetch the paper's landing page / official abstract.
3. Extract the essentials:
   • 核心贡献 Contribution — the single new idea in one sentence
   • 方法 Method — high-level approach, 2–3 bullets, no dense math
   • 关键结果 Results — headline numbers / findings
   • 局限 Limitations — what the authors acknowledge (from abstract or conclusion)
   • 影响 Impact — why it matters / who cares
4. Output bilingual — every section has both 中文 and English lines.
5. If the abstract is unavailable or the page cannot be fetched, say so plainly and do not guess.

Output format | 输出格式

# 📄 <Paper title> | 论文要点
- **arXiv / 链接**: https://arxiv.org/abs/<id>
- **核心贡献 Contribution**: 一句话 | one sentence
- **方法 Method**:
  - ··· | ···
- **关键结果 Results**: ··· | ···
- **局限 Limitations**: ··· | ···
- **一句话点评 TL;DR**: 值得精读吗？| worth a full read? yes/no + why

Tips | 提示

• Prefer the abstract page over a PDF; PDFs are slower and harder to parse reliably.
• Do not fabricate numbers — only state results that are actually in the abstract/text.
• If the user wants depth (method details), offer to fetch the full HTML version and expand.
• Keep the summary under ~250 words when the user just asks for 要点.