We have developed one of the most self-aware LLMs that knows about itself in the domain of Retrieval-Augmented Generation (RAG). With over 1,400 papers on RAG stored in our vector database, our LLM acts as a RAG expert, capable of answering almost any question on the subject by retrieving and utilizing relevant data. This makes it a valuable resource for researchers, students, and professionals interested in RAG.
Data🗂️
Our data is sourced from arXiv, a leading platform for academic papers, with a focus on papers related to RAG. These papers are scraped, processed, and stored in Pinecone's vector database as embeddings. When a user asks a question, the system retrieves the most relevant papers or sections, and uses them as context for generating accurate, context-driven answers.
Benefit🎯
Comprehensive Expertise: Provides detailed, research-based insights on RAG.
Rapid Access: Users can retrieve information from over 1,400 papers in real time, making research more efficient.
Innovative Learning: Ideal for learning, exploring, and deepening knowledge in the field of RAG through a conversational interface.
Research Support: Helps researchers by reducing the time spent searching through large bodies of work.
Architecture and Techinal🚀
Architecture of AskmeAboutRAG
example of 1000 datapoints visualize from vector database
Project Name
AskmeAboutRAG
Description✨
We have developed one of the most self-aware LLMs that knows about itself in the domain of Retrieval-Augmented Generation (RAG). With over 1,400 papers on RAG stored in our vector database, our LLM acts as a RAG expert, capable of answering almost any question on the subject by retrieving and utilizing relevant data. This makes it a valuable resource for researchers, students, and professionals interested in RAG.
Data🗂️
Our data is sourced from arXiv, a leading platform for academic papers, with a focus on papers related to RAG. These papers are scraped, processed, and stored in Pinecone's vector database as embeddings. When a user asks a question, the system retrieves the most relevant papers or sections, and uses them as context for generating accurate, context-driven answers.
Benefit🎯
Architecture and Techinal🚀
Architecture of AskmeAboutRAG
example of 1000 datapoints visualize from vector database
Technology & Languages
Project Repository URL
https://github.com/SupeemAFK/AskMeAboutRAG-Main
Deployed Endpoint URL
https://huggingface.co/spaces/Supeem/Askme_About_RAG
Project Video
https://www.youtube.com/watch?v=pBBGc7fE42U
Team Members
SupeemAFK, microhum, beambeambeam, GGital