A reading note, not a recap
The note preserves the paper's question, method backbone, key results, limitations, and reusable research context.
Turn one dense paper into an Obsidian note worth keeping.
DeepPaperNote is an evidence-first reading workflow for researchers who want more than an abstract summary.
The output is the product.
DeepPaperNote is built around the note you keep: structured, evidence-aware, figure-conscious, and ready for a long-term Obsidian vault.
Claims follow the source
Metadata, sections, captions, formulas, and numeric claims are gathered before the final synthesis begins.
Figure-aware by default
Major figures, tables, and formulas stay attached to the reasoning they support, so visual evidence remains part of the deep read.
Generated structure
DeepPaperNote output
- 01 原文摘要翻译
- 02 创新点
- 03 方法主线
- 04 关键结果
- 05 深度分析
Visual evidence
Figures, tables, and formulas stay with the claims they support.
DeepPaperNote keeps visual and mathematical evidence inside the Obsidian note, close to the reasoning built from it.
Fig. 2 Model architecture overview caption matched
Input tokens
Embedding
Multi-head attention residual + norm
Feed-forward
Output states
Attention(Q, K, V) = softmax(QKT / √dk)V
| Evidence | Source | Role |
|---|---|---|
| Architecture diagram | Fig. 2 | Model flow |
| Attention equation | Eq. 1 | Core math |
| Ablation table | Table 4 | Result check |
A disciplined path from paper to note.
Scripts prepare structured evidence. The model does the reading. The final note is linted, reviewed, and saved where your research actually lives.
- ResolvePick one canonical paper identity.
- EvidenceExtract sections, captions, metrics, and context.
- FiguresPlan major figure and table placeholders first.
- SynthesisTurn the bundle into a deep-reading note.
- LintCheck structure, style, and note completeness.
- SaveWrite to an Obsidian-style paper folder.
Install the skill. Hand it one paper.
Keep the website out of the skill package. Keep the workflow focused on the note.
npx skills add 917Dhj/DeepPaperNote python3 -m pip install PyMuPDF Use DeepPaperNote on this paper: https://arxiv.org/abs/1706.03762