AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Academic Plotting for ML Papers

Turns paper drafts or raw experiment results into publication-quality architecture diagrams and conference-sized data charts.

Data & AnalyticsIntermediate★ 1,337⑂ 91AI score 9/10Last updated: Oct 9, 2026

What it does

  • Routes to the right workflow: numerical axes → matplotlib/seaborn; boxes and arrows → Gemini image generation.
  • Extracts from context: reads a method section, a descriptive paragraph, a results table, or CSV/JSON and pulls out components, relationships, methods, and metrics.
  • Auto-selects chart type: time/step axis → line plot; N methods × M benchmarks → grouped bars; square matrix → heatmap; single ranking → horizontal bar leaderboard.
  • Publication styling baked in: serif fonts, 300 DPI, despined axes, colorblind-safe Okabe-Ito palette, a distinct highlight color for "Ours", and dual PDF (vector) + PNG export.
  • Venue size table: single-column and full-width inch specs for NeurIPS, ICML, ICLR, ACL, AAAI.
  • Four diagram styles: hand-drawn sketch, modern minimal, illustrated technical, classic accent bar — each with a full prompt block down to hex codes.
  • Reproducibility: always saves the generator at figures/gen_fig_<name>.py with a strict file-naming convention.

Who it's for

  • ML/AI researchers and grad students submitting to NeurIPS, ICML, ICLR, or ACL
  • Anyone who burns hours on the Figure 1 overview diagram
  • Teams with experiment logs but no consistent chart style across a paper
  • Authors who have been told their figures look unpolished by reviewers

Examples

  1. Architecture diagram: "Our system has a Planner, Executor and Verifier; the Verifier feeds back to the Planner on failure" → cycle layout with three nodes and a dashed feedback arrow, generated three times with Gemini so you can pick the best.
  2. Comparison bar chart: "GPT-4: MMLU 86.4 / HumanEval 67.0; Ours: 88.1 / 71.2; Llama-3: 79.3 / 62.1" → grouped bars for 3 methods × 2 benchmarks, "Ours" in coral, value labels on bars, exported to PDF.
  3. Training curves: a CSV of per-step accuracy mean/std → line plot with markers and shaded confidence bands, sized to ICML's 3.25-inch single column.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/OpenRaiser/NanoResearch/HEAD/skills/vendor-ai-research/academic-plotting/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the skills/vendor-ai-research/academic-plotting folder from the GitHub repo OpenRaiser/NanoResearch into my ~/.claude/skills/academic-plotting/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/OpenRaiser/NanoResearch.git /tmp/NanoResearch && mkdir -p ~/.claude/skills && cp -r /tmp/NanoResearch/skills/vendor-ai-research/academic-plotting ~/.claude/skills/ && rm -rf /tmp/NanoResearch

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and make sure the skills folder exists: mkdir -p ~/.claude/skills
  2. Clone the repository: git clone https://github.com/OpenRaiser/NanoResearch.git /tmp/NanoResearch
  3. Copy just this skill: cp -r /tmp/NanoResearch/skills/vendor-ai-research/academic-plotting ~/.claude/skills/
  4. Clean up: rm -rf /tmp/NanoResearch
  5. Install the Python dependencies: pip install "matplotlib>=3.8" "seaborn>=0.13" numpy "google-genai>=1.0"
  6. (Only needed for AI diagram generation) Get a Gemini key at https://aistudio.google.com/apikey and export it: export GEMINI_API_KEY="your_key"
  7. Restart Claude Code, then ask something like "Make an ICML-ready figure from these results."
  8. Outputs land in your project's figures/ folder as gen_fig_*.py, fig_*.pdf, and fig_*.png.
View source on GitHub ↗License: MIT