Hi everyone,
I recently created a new app called Simple Papers, and I’m very excited to share it with you all.
Over the years, I’ve tried to read and digest many landmark ML/NLP and lately LLM papers and ended up abandoning them because I either got bogged down by the dense content, or got lost because I was missing some fundamentals.
Simple Papers is my attempt to fix that. It turns each section of an academic paper into a short, digestible explanation, so you can actually go through the paper as you normally would, but understand the core ideas without getting lost in the details.
“It doesn’t just summarize the paper, it simplifies it for you!” (marketing tagline still in beta!)
I had a ton of fun building this, stitching together a bunch of my favorite tools, and learned a lot from the papers it simplified for me. Hope the app helps you too whether you use it to understand the papers, or use the code as inspiration for your next awesome project.
Check out this silly trailer to learn more about the app:
Tech stack used:
- Landing AI’s Agentic Document Extraction for parsing PDFs
- Anthropic’s Sonnet 3.7 and 4.5 for simplifications
- Amazon Web Services (AWS) Bedrock + LangChain for model serving and orchestration
- ElevenLabs for text-to-speech
- Streamlit for the UI (
of course!) - And pen and paper for the logo (touched up by AI cuz it’s 2025)!
Huge thanks to @thiago, Shobeir Seddington, @dataprofessor, Tyler Richards, @SiddhantSadangi and the rest of the Streamlit creators for their awesome feedback on the initial drafts of the app.
Check out the app and let me know what you think!