explaingit

hkuds/paper2slides

3,625Python
This is a quick first-pass explanation. The richer sections — use-cases, tech stack, setup, prompts — are still being generated.

TLDR

Paper2Slides is a tool that takes a research paper, report, or other document and turns it into a presentation slide deck or a poster automatically.

Mindmap

A visual breakdown will appear here once this repo is fully enriched.

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

In plain English

Paper2Slides is a tool that takes a research paper, report, or other document and turns it into a presentation slide deck or a poster automatically. You give it a file and it produces ready-to-use slides without you having to build anything from scratch. It works from the command line or through a local web interface you can run in your browser. The tool supports a wide range of input file types: PDF, Word documents, Excel spreadsheets, PowerPoint files, and Markdown. You can feed it multiple files at once. Internally it runs through four stages: it first parses the document and builds a searchable index of its content (using a technique called RAG, which lets it find and quote specific parts accurately), then it extracts the structure and key figures, then it plans a layout, and finally it renders the slides or poster. You have control over the output style. There are built-in themes with names like "academic" and "doraemon," but you can also describe a style in plain English and the tool will try to match it. Output length is configurable too, with short, medium, and long options for slide decks, and sparse, medium, or dense options for posters. One practical feature is the checkpoint system. The tool saves its progress at each stage, so if you stop it partway through, you can resume from where it left off by running the same command again. You can also force it to restart from a specific stage, which is useful if you want to change the style without re-parsing the document. The project requires Python 3.12 or newer, and you need to supply your own AI API keys in a configuration file. Parallel generation is available to speed up processing. A web interface can be started with a single script and is then accessible at a local address in your browser.

Open on GitHub → Explain another repo

← hkuds on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.