paper-summary

Name: paper-summary
Availability: InStock
Author: collat-labs

Extract evidence-grounded research problem, methodology, contributions, results, and limitations from academic papers.

1|Updated Apr 25, 2026

Search Tags:#pdf #arxiv #evidence extraction #paper summary #academic analysis #chinese output #hallucination control

Ready to run

npx skills add https://github.com/collat-labs/collat --skill paper-summary-collat-labs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: paper-summary
Source: https://github.com/collat-labs/collat/tree/main/.agents/skills/paper-summary
Command: npx skills add https://github.com/collat-labs/collat --skill paper-summary-collat-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

帮助你在阅读论文时快速获得可核查的研究要点，避免把“看起来像对的内容”当成事实，从而更高效地完成文献理解、筛选与对比。

Core Features & Use Cases

结构化抽取关键学术信息：从论文中提取研究问题、方法、关键贡献、实验结果（含量化数字）与局限性，且在信息缺失时明确标注为未说明。
证据对齐与可追溯性：对每个主要结论提供来自论文的直接证据，并尽可能给出位置线索（段落/表格/图等）。
质量评估与自我修正：进行内部自检（不向用户暴露分数 JSON），当质量不足或可能幻觉风险较高时进行最多两轮的定向精炼。
适用场景：用于科研助理式阅读、课程作业/中文输出的学生材料准备、以及需要“带证据”的论文综述与对比。

Quick Start

使用 $paper-summary 分析你提供的 arXiv 链接或本地 PDF 文件，并生成中文、证据对齐且不输出评分 JSON 的论文结构化总结。

Frequently Asked Questions about paper-summary

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I extract key findings from an arXiv paper without hallucinations?▼

To extract key findings from an arXiv paper without hallucinations, use a structured summarization approach that demands source-only evidence. This Skill grounds each claim directly in the text with location markers, ensuring reliable academic reporting.

Can I summarize a local PDF for a Chinese academic literature review?▼

Yes, you can summarize a local PDF for a Chinese academic literature review. The Skill processes local PDF inputs and arXiv links to output structured Chinese summaries covering research problems, methodology, and experimental results.

What is the best way to extract methodology and quantitative results from an academic paper?▼

The best way to extract methodology and quantitative results is using an evidence-grounded extraction tool. It identifies specific research methods and quantitative numbers while explicitly marking any missing details as unreported to maintain accuracy.

Does the paper summary tool support batch processing for multiple PDFs?▼

No, the paper summary tool does not support batch processing for multiple PDFs. It is designed for single academic paper analysis to ensure strict source-only extraction and high-quality evidence alignment without compromising output reliability.

How does evidence extraction handle missing information or potential hallucinations?▼

Evidence extraction handles missing information by inserting mandatory fallback text indicating details are unreported. It also runs internal self-evaluations to detect high hallucination risks, triggering up to two rounds of targeted refinement.

When do I need evidence-aligned location markers for academic analysis?▼

You need evidence-aligned location markers when preparing literature reviews or relevance screening materials. Providing paragraph, table, or figure references ensures every extracted contribution and limitation is verifiable directly against the original paper.