PolarisLiu1/LAT

Look As You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning (Poster of AAAI'26)

23
/ 100
Experimental

This project helps make information retrieved from documents more trustworthy and easier to verify. It takes documents (like PDFs or web pages) and user questions, then produces answers alongside clear, step-by-step reasoning that points directly to the exact visual evidence in the original documents. This is for professionals who need highly reliable and explainable answers from large document sets, like researchers or legal analysts.

Use this if you need to understand not just the answer to a question from a document, but also the detailed reasoning and specific visual evidence that supports each part of that answer.

Not ideal if you only need quick answers without detailed evidence or if your documents are purely text-based without any visual elements that need to be referenced.

document-analysis information-retrieval research-verification evidence-gathering explainable-AI
No License No Package No Dependents
Maintenance 6 / 25
Adoption 4 / 25
Maturity 5 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

Python

License

Last pushed

Dec 01, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/PolarisLiu1/LAT"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.