HKUDS/Paper2Slides
Paper2Slides
Paper2Slides is an AI-driven presentation generator and content extractor designed to transform academic papers and scientific documents into structured slides and posters. It utilizes retrieval-augmented generation to distill key data points and identify critical figures while maintaining direct traceability to the original source text.
The system functions as an AI slide designer that applies professional themes or custom visual styles defined through natural language. It integrates with external image generation services to produce high-quality visuals and research visualizations for academic layouts.
The application employs a multi-stage generation pipeline that supports both deep semantic indexing for complex papers and a fast-mode for rapid previews. Processing efficiency is managed through parallel worker rendering and a state-based checkpoint system that allows for the resumption of interrupted tasks or the independent regeneration of specific sections.
The project is provided as a containerized document processor to ensure consistent execution across different operating systems.
Features
- Presentation Generators - Transforms research papers and academic documents into professional presentation slides using AI-driven extraction and summarization.
- RAG Document Retrieval - Uses retrieval-augmented generation to distill and retrieve key data snippets from documents for presentation content.
- Content Extractors - Implements a pipeline that identifies critical data and figures from source documents with direct traceability for verification.
- Retrieval-Augmented Generation - Employs retrieval-augmented generation to ground output content in the source text, ensuring traceability and accuracy.
- Document-to-Presentation Converters - Transforms academic papers and various file formats into structured presentation slide decks.
- Content Source Traceability - Maintains direct links between output presentation elements and the original source document for verifiability.
- Research Presentation Generation - Automatically produces academic presentations and slide decks by extracting and summarizing findings from research papers.
- Multi-Stage Pipeline Processing - Utilizes a multi-stage pipeline to transform raw document text into structured presentation layouts through extraction and summarization.
- Generative AI Insights - Uses generative AI to extract critical data points and figures from documents while maintaining source traceability.
- Scientific Document Processing - Processes complex scientific documents and academic papers to transform them into structured visual formats.
- Automated Slide Generators - Functions as an AI designer that automates the creation of professional slide visuals and layouts.
- Image Generation Integrations - Integrates external AI-powered image synthesis services to generate high-quality research figures and visuals.
- Adaptive Indexing - Optimizes processing speed by allowing a choice between deep semantic indexing for complex papers and a fast mode for short files.
- Research Poster Generation - Transforms academic documents into structured poster formats by distilling key findings and data from source text.
- Research Figure Generation - Produces high-quality research figures and visualizations for academic presentations via integrated image generation.
- Parallel Rendering - Distributes the slide rendering process across multiple concurrent workers to increase the overall production speed.
- Visual Style Customization - Applies professional themes and custom visual styles to presentations based on natural language descriptions.
