PaperLens — Understand Research Papers, Faster.
An open-source AI-powered research assistant that helps you query, explore, and reason over your paper collection using retrieval-augmented generation.
Research Workflow, Distilled
PaperLens keeps the paper at the center of the conversation. Upload once, ask repeatedly.
Ingest
PDFs are indexed with semantic embeddings.
Retrieve
Relevant sections are surfaced for each query.
Reason
Answers stay anchored to cited paper context.
Reading papers shouldn't break your flow.
Research papers are dense, long, and time-consuming. Searching PDFs manually interrupts the thinking process, and existing tools are often generic or closed-source.
Notes, highlights, and cross-paper reasoning get fragmented across tabs and tools. PaperLens pulls those threads into one place so you can stay focused on the work that matters: understanding the research.
Core Friction
- Manual PDF searches derail deep reading sessions.
- Answers aren't grounded or traceable in the source paper.
- Paper collections grow faster than researchers can keep up.
A research-focused assistant built around RAG.
PaperLens is designed specifically for papers, not general chat. It is self-hosted, extensible, and transparent so you always know where answers come from.
Upload papers and build a searchable library.
Papers are indexed and embedded for semantic retrieval.
Ask natural language questions as you read.
Answers are grounded in the paper content, not guesswork.
What works today
Focused capabilities for researchers who want grounded answers without vendor lock-in.
Paper Ingestion & Indexing
Upload and process PDFs with automatic chunking and embedding for retrieval.
Semantic Search
Ask questions in natural language and retrieve relevant paper sections.
Context-Aware Question Answering
Responses are grounded in retrieved paper context to reduce hallucinations.
Chat-Based Research Workflow
Iterative, conversational exploration with session history.
Local / Self-Hosted Friendly
Run locally or on your own server with no forced vendor lock-in.
Developer-Focused Architecture
FastAPI-based, modular design for embeddings, vector stores, and LLMs.
On the roadmap
The next phase focuses on personalization, richer citations, and a research-first workspace.
Multi-Paper Reasoning
Ask questions that span multiple papers and compare results.
Section-Aware Queries
Target specific sections like Method, Results, or Limitations.
Personalized Ingestion Pipeline
Ingest papers based on user-defined research interests to boost personalization.
Daily Research Digest Reports
Generate daily overviews of newly ingested papers for quick scanning and triage.
Paper Annotations & Notes
Save highlights, summaries, and personal research notes.
Better Citation Tracing
See exactly which paper chunks contributed to each answer.
User Accounts & Persistence
Authentication, saved sessions, and persistent paper libraries.
Improved Research Dashboard
Paper-centric navigation with a cleaner reading experience.
Built in public and evolving fast.
PaperLens is open-source and built as a learning + research-driven project. Issues and pull requests are welcome, and the roadmap is shaped by real researcher workflows.
Why open source?
Research tooling should be transparent, inspectable, and easy to extend.
Star the repo to support the project.
Follow progress, see what's shipping, and help define what comes next.
Get running locally.
PaperLens is designed to run on your own machine or server. Start with the README to configure your environment and launch with Docker or FastAPI.
Clone the repository from GitHub.
Follow the README to configure your environment.
Run locally with Docker or directly via FastAPI.
Questions, bugs, or collaboration?
Reach out for feature requests, partnership ideas, or research collaboration.
Contact & Support
We respond fastest through GitHub Issues.
Open Issues
Bug reports and feature requests
Send Email
Direct line for collaboration
We keep links compact here to reduce visual clutter.
About the builder
I'm Hridaya Sharma, the creator of PaperLens — a final year ML undergrad with hands-on machine learning and research experience. I build open-source research tools that keep the paper at the center of the workflow.
Connect
Personal links for updates and collaboration.
Hridaya Sharma
ML undergrad · Research builder