RehgLab

RehgLab / ArcDeck

Public

Narrative-Driven Paper-to-Slide Generation via ArcDeck

arcdeck.org
43
0
100% credibility
Found Apr 29, 2026 at 43 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

ArcDeck automatically generates professional PowerPoint presentations from academic PDF papers by analyzing narrative structure, matching visuals, and applying themes.

How It Works

1
📚 Pick your research paper

Choose a PDF of an academic paper you want to turn into presentation slides.

2
💻 Open ArcDeck

Launch the simple web app that feels like using a photo editor.

3
📤 Upload the PDF

Drag your paper file into the app and let it read the content.

4
⚙️ Set talk details

Tell it how long your talk is and who your audience is, like students or experts.

5
🎨 Pick a theme

Choose a beautiful color style from previews, like ocean blue or warm sunset.

6
✨ Generate slides

Click go and watch ArcDeck understand the paper's story, match pictures, and build polished slides.

7
📥 Download PPTX

Get your ready-to-present PowerPoint file with speaker notes if you want.

🎉 Present with confidence

Your slides tell the paper's story perfectly, impressing everyone.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 43 to 43 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ArcDeck?

ArcDeck automates narrative-driven paper-to-slide generation via Python, transforming academic PDFs into polished PPTX decks. It parses the paper's discourse structure to create coherent arc decks that follow the story's flow, complete with matched figures, tables, speaker notes, and themes. Run it via CLI (`python -m slide_generation.arcdeck paper.pdf --model gpt-4o --duration 20`), Gradio web UI, or even Claude skills for quick conference-ready output.

Why is it gaining traction?

Unlike generic summarizers, ArcDeck uses RST discourse trees and a multi-agent refinement loop for logical slide plans that respect the paper's narrative spine—think better flow without manual tweaking. It handles visuals smartly (aspect-preserving placement, up to 12 templates or JS designs), plus ArcBench benchmark for eval. Early adopters love the end-to-end polish from arXiv PDF to themed deck.

Who should use this?

ML researchers prepping NeurIPS/ICLR orals from fresh papers, PhD students rushing slide decks for quals, or profs generating arc decks for seminars. Ideal if you're tired of copy-pasting figures and rewriting bullets manually.

Verdict

Grab it for academic talks—solid CLI/UI and benchmark make it practical despite 43 stars and 1.0% credibility signaling early days. Docs are thorough (arXiv paper + HF dataset), but expect tweaks for edge-case PDFs; test on your papers first.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.