Paper Reading

Step 1: PDF Acquisition

All papers are processed as PDF. No HTML/ar5iv path.

Download Flow

# For arXiv: extract ID and download PDF
curl -L -o <output_dir>/paper.pdf "https://arxiv.org/pdf/XXXX.XXXXX"

# For other URLs: try direct download
curl -L -o <output_dir>/paper.pdf "<url>"
# Verify it's a valid PDF: file <output_dir>/paper.pdf should show "PDF document"

Read Content

Use the Read tool to read the PDF file. Claude natively supports reading PDF files and extracting text content. For large PDFs (>10 pages), read in page ranges (e.g., pages: "1-10", then pages: "11-20").

Step 2: Paper Type Identification

After reading the title, abstract, and introduction, determine paper type:

When uncertain, default to the Empirical template.

Step 3: Figure & Table Extraction (pymupdf4llm)

1. Prepare Output Directory

mkdir -p <output_dir>/images

2. Screenshot Priority Guide

Capture 3-8 key figures per paper. If the paper has few or no meaningful figures (e.g., theoretical papers, short workshop papers), skip figure extraction and produce a text-only summary.

3. Automated Extraction with pymupdf4llm

Use pymupdf4llm to extract all images and vector graphics in one call. This handles raster images (photos, embedded figures) AND vector graphics (plots, diagrams, flowcharts) automatically.

Prerequisites: pymupdf and pymupdf4llm must be installed in the conda base environment.

# Install if needed (one-time)
source $HOME/anaconda3/etc/profile.d/conda.sh && conda activate base
pip install pymupdf4llm

Run extraction:

source $HOME/anaconda3/etc/profile.d/conda.sh && conda activate base && python3 << 'PYEOF'
import pymupdf4llm
import os

pdf_path = "<pdf_path>"
image_dir = "<output_dir>/images"
os.makedirs(image_dir, exist_ok=True)

# Extract all figures, tables, and vector graphics as images
md_result = pymupdf4llm.to_markdown(
    pdf_path,
    write_images=True,
    image_path=image_dir,
    image_format="png",
    dpi=200,
    use_ocr=False,        # Disable OCR (avoids tesseract dependency)
)

# List extracted images
for f in sorted(os.listdir(image_dir)):
    if f.endswith(('.png', '.jpg', '.jpeg')):
        from PIL import Image
        try:
            img = Image.open(os.path.join(image_dir, f))
            print(f"{f}: {img.size[0]}x{img.size[1]}")
        except:
            print(f"{f}: (size unknown)")
PYEOF

4. Filter & Rename Extracted Images

pymupdf4llm extracts ALL graphical regions, including logos, watermarks, and decorative elements. Filter and rename:

Step 4a: Filter out noise — Remove images that are:

• Too small (< 100x100 px) — likely icons or logos
• Too narrow (aspect ratio > 10:1 or < 1:10) — likely decorative lines or borders

source $HOME/anaconda3/etc/profile.d/conda.sh && conda activate base && python3 << 'PYEOF'
import os
from PIL import Image

image_dir = "<output_dir>/images"
removed = []
for f in os.listdir(image_dir):
    path = os.path.join(image_dir, f)
    try:
        img = Image.open(path)
        w, h = img.size
        ratio = max(w, h) / max(min(w, h), 1)
        if w < 100 and h < 100:
            os.remove(path)
            removed.append(f"{f} (too small: {w}x{h})")
        elif ratio > 10:
            os.remove(path)
            removed.append(f"{f} (too narrow: {w}x{h})")
    except:
        pass

print(f"Removed {len(removed)} noise images:")
for r in removed:
    print(f"  - {r}")

kept = [f for f in sorted(os.listdir(image_dir)) if f.endswith(('.png', '.jpg'))]
print(f"\nKept {len(kept)} images:")
for f in kept:
    img = Image.open(os.path.join(image_dir, f))
    print(f"  {f}: {img.size[0]}x{img.size[1]}")
PYEOF

Step 4b: Visual review & rename — Use the Read tool to view each remaining image. Based on content:

• Rename to descriptive names: figure_1_overview.png, table_2_main_results.png, etc.
• Delete any remaining non-figure images (headers, footers, etc.)
• Select 3-8 key figures for the summary based on the priority guide above

5. Fallback: Manual pymupdf clip extraction

If pymupdf4llm misses a specific figure or table (rare), use direct pymupdf clip rendering:

source $HOME/anaconda3/etc/profile.d/conda.sh && conda activate base && python3 << 'PYEOF'
import fitz

doc = fitz.open("<pdf_path>")
page = doc[PAGE_NUM]  # 0-indexed

# Render a specific region at high resolution
clip = fitz.Rect(x0, y0, x1, y1)  # coordinates from page layout
mat = fitz.Matrix(3, 3)  # 3x zoom
pix = page.get_pixmap(matrix=mat, clip=clip)
pix.save("<output_dir>/images/figure_N_desc.png")
doc.close()
PYEOF

To find coordinates: use page.get_text("dict") to find text blocks containing "Figure N" or "Table N", then estimate the figure region nearby.

File Naming

Format: figure_N_<brief_desc>.png / table_N_<brief_desc>.png

Examples:

• figure_1_overview.png — system overview
• figure_2_architecture.png — model architecture
• table_3_comparison.png — main results table
• figure_4_ablation.png — ablation study

Step 4: Fill Template

After identifying paper type, select the corresponding template

Writing Principles (Critical)

Depth-first: For every section, ask "why" and "how", not just "what".

After identifying paper type, select the corresponding template

All types share these sections:

## Basic Information
- **Title:**
- **Authors:**
- **Affiliation:** (optional)
- **Published:**
- **Link:**
- **Paper Type:** [Empirical / Theoretical / Survey / Systems]
- **One-line summary:** What was done + how + what was the result

## Research Problem
- **What problem does it solve?** Identify the specific gap in existing methods
- **Key assumptions:** What constraints/limitations frame the research
- **Why is it important?** The practical impact on the field
- **Positioning among related work:** What are the 2-3 closest prior works? What is the key difference?

Template A: Empirical Paper

## Basic Information
[shared section]

## Research Problem
[shared section]
- **Mathematical formulation:** (optional)

<!-- Insert problem definition/motivation figure here if available -->

## Key Insight
> Distill the paper's core new idea in 2-3 sentences. Not "what was done", but "what insight makes this method work".
> Example: Rather than predicting frame-by-frame, first establish long-term 3D point tracking, then leverage temporal consistency for joint optimization.

## Technical Method
### Overall Framework and Principles
<!-- Insert architecture diagram -->
<!-- Insert figure here: ![Figure N: description](./images/figure_N_desc.png) -->

- Overall system architecture description
- Modules/components and their responsibilities
- Signal/data flow direction
- **Why this design?** Advantages over the intuitive/naive approach

### Core Component Details
<!-- Insert algorithm flowchart here if available -->

- Model/algorithm architecture details (layers, dimensions, input/output)
- Training objective and loss function (write key equations)
- Training data source (synthetic/real/mixed, dataset names and scale)
- Key tricks and design decisions
- **Motivation for each design choice:** Why use A instead of B? Does the paper provide justification?

## Experimental Results
<!-- Insert experiment result figures/tables -->
<!-- Insert figure here: ![Figure N: description](./images/figure_N_desc.png) -->

### Results (Facts)
- **Experimental setup:** Environment, hardware, hyperparameters
- **Baselines compared:** List specific method names and sources
- **Key results:** Quantitative improvement margins (specific numbers + percentages)
- **Ablation study:** Component contributions (removing X decreases performance by Y%)
- **Surprising findings:** Any counterintuitive results

### Analysis (Interpretation)
- Authors' explanation and attribution of results
- Which scenarios/datasets show best performance? Worst?
- Root cause of performance gains (authors' claims vs actual evidence)

<!-- Insert ablation or visualization result figures here if available -->

## Critical Analysis
### Strengths
- Specific improvements over prior work (not just "good results")

### Limitations
- **Acknowledged by authors:**
- **My observations:** Issues not mentioned in the paper
  - Do assumptions hold in practice?
  - Are compute/data requirements reasonable?
  - Are evaluation metrics comprehensive?

### Reproducibility Assessment
- Is code open-sourced? Is data available?
- Are key implementation details sufficiently described?

## Summary
[shared section]

Template B: Theoretical Paper

## Basic Information
[shared section]

## Research Problem
[shared section]
- **Mathematical formulation:** Formal problem definition

## Key Insight
> Distill the paper's core theoretical contribution in 2-3 sentences. What new mathematical tool/perspective makes this result possible?

## Theoretical Framework
### Problem Formalization
- Symbol definitions and notation conventions
- Core mathematical definitions

### Main Theorems and Proof Sketches
- **Theorem 1:** Statement + key proof idea (not full proof, but key steps and key lemmas)
- **Theorem 2:** ...
- Key techniques used in proofs: Why does this technique work? Is there a more intuitive explanation?

### Theoretical Analysis
- Implications and intuitive interpretation of results (restate in non-mathematical language)
- Tightness of upper/lower bounds
- Relationship to and comparison with known results: Which bound was improved? Which assumption was relaxed?

## Validation (if experiments exist)
- Experimental setup
- Comparison of theoretical predictions vs actual results
- How is the gap between theory and experiments explained?

## Critical Analysis
### Strengths
- Importance and novelty of the theoretical contribution

### Limitations
- **Acknowledged by authors:**
- **My observations:** Reasonableness of assumptions, practical utility, difficulty of generalization

## Summary
[shared section]

Template C: Survey Paper

## Basic Information
[shared section]
- **Coverage:** Number of papers surveyed, time span

## Research Problem
[shared section]

## Key Insight
> What is the core contribution of this survey? What classification perspective was proposed, or what important trends were identified?

## Taxonomy
<!-- Insert taxonomy/classification figure -->
<!-- Insert figure here: ![Figure N: description](./images/figure_N_desc.png) -->

- Main classification dimensions and rationale for their selection
- Category definitions and representative works

### Direction 1: [Name]
- Key methods and advances
- Representative works (author, year)
- Pros and cons
- **Current bottleneck:** The core challenge facing this direction

### Direction 2: [Name]
- ...

### Method Comparison
| Method Type | Strengths | Weaknesses | Representative Works | Best Use Case |
|-------------|-----------|------------|---------------------|---------------|
| ... | ... | ... | ... | ... |

## Open Problems and Trends
- Current major challenges in the field
- Emerging trends and directions
- Authors' predictions and recommendations
- **Most promising direction:** Based on the survey analysis, which direction deserves most attention? Why?

## Critical Analysis
- Is the survey's coverage comprehensive? Any important directions missed?
- Is the taxonomy reasonable? Could it be organized better?
- Do the authors' opinions/biases affect the survey's objectivity?

## Summary
[shared section]

Template D: Systems Paper

## Basic Information
[shared section]

## Research Problem
[shared section]
- **Design goals:** Key requirements the system must meet

## Key Insight
> What is the core design insight of this system? What trade-off or observation makes this design superior to existing solutions?

## System Design
### Architecture Overview
<!-- Insert system architecture diagram -->
<!-- Insert figure here: ![Figure N: description](./images/figure_N_desc.png) -->

- Overall architecture and component breakdown
- Component responsibilities and interfaces

### Key Design Decisions
- Decision 1: What choice was made, why (and not the alternative)
- Decision 2: Trade-offs considered (performance vs complexity vs maintainability)
- Key differences from existing systems

### Implementation Details
- Key tech stack/dependencies
- Optimization techniques
- Fault tolerance / scalability design

## Performance Evaluation
<!-- Insert performance comparison figures/tables -->
<!-- Insert figure here: ![Figure N: description](./images/figure_N_desc.png) -->

### Experimental Facts
- **Benchmark setup:** Environment, hardware, workloads
- **Compared systems:**
- **Key metrics:** Throughput, latency, resource usage (specific numbers)
- **Scalability:** Performance as scale increases

### Result Interpretation
- Under what conditions does it perform best? When does it degrade?
- Root cause of performance advantages

## Deployment Experience (if available)
- Real-world production performance
- Problems encountered and solutions

## Critical Analysis
### Strengths
- Design elegance and practicality

### Limitations
- **Acknowledged by authors:**
- **My observations:** Deployment assumptions, hardware dependencies, generalizability

## Summary
[shared section]

Shared Summary Section

## Summary and Evaluation

### Three-Perspective Conclusion (Andrew Ng Framework)

**Authors' conclusion:** What do the authors claim to have accomplished? What goals were achieved?

**Personal assessment:**
- Did the authors truly achieve their claimed goals? Is the evidence sufficient?
- How much does this work actually advance the field?

**Overall evaluation:**
- **Core idea:** One sentence summarizing the core contribution
- **Main highlight:** What stands out compared to prior work
- **Future directions:** Natural next steps
- **Rating:** [Breakthrough / Important / Valuable / Incremental]

### Comprehension Verification (Self-check after writing)
1. What were the authors trying to accomplish?
2. What are the key elements of the approach?
3. What can I use in my own research?
4. What references are worth reading further?

Section Writing Guidelines

Basic Information

• Extract from paper header, abstract, or metadata
• For links: use DOI if available, otherwise arXiv or publisher URL
• One-line summary must include: what was done + how + what was the result

Research Problem

• Focus on the specific gap the paper addresses, not generic field descriptions
• Mathematical description: include key equations if present
• Assumptions: what constraints or simplifications does the approach make?
• Related work positioning: Must mention 2-3 closest prior works and explain what makes this paper unique

Key Insight

• This is the most important paragraph — distill the key insight that makes the method work
• Not a paraphrase of the abstract, but answering "what is this paper's core new idea?"
• Good example: "Observed phenomenon X, thus leveraging Y to achieve Z"
• Bad example: "Proposes a new method to solve problem A" (too vague)

Technical Method (Empirical)

• Architecture first: Draw the big picture before diving into components
• Be specific: Dimensions, layer counts, parameter counts — not just "a network"
• Loss functions: Write the actual LaTeX equation if provided
• Training data: Note if synthetic, real-world, or mixed; mention dataset names and scale
• Design motivation: Every key design choice should explain why (paper's justification or inferred reasoning)
• Insert architecture diagram screenshot if available

Experimental Results

• Separate facts from interpretation: Results section writes experimental data only; Analysis section writes explanations and attribution
• Focus on quantitative improvements over baselines (specific numbers and percentages)
• Note which metrics matter most for this problem domain
• Mention any surprising or counterintuitive results
• Note best and worst performing scenarios/datasets
• Insert main result figure/table screenshot if available

Critical Analysis

• Strengths: Must specify concrete improvements, not just "good results"
• Limitations: Separately list author-acknowledged and self-discovered ones
• Reproducibility: Is code/data open-sourced? Are implementation details sufficient?

Summary and Evaluation

• Three perspectives: Authors' conclusion → Personal assessment → Overall evaluation
• Rating: Choose from [Breakthrough / Important / Valuable / Incremental] with justification
• Comprehension verification: Self-check with 4 questions after writing (Andrew Ng framework)

Common Mistakes

Language

• Output summary in the user's preferred language
• Technical terms can remain in English (API, Loss, Baseline, etc.)
• Code and equations in original form
• Translate figure captions to user's preferred language