This repository contains two key projects:
An intelligent multi-agent system that helps researchers search, analyze, explain, and summarize academic papers using Large Language Models (LLMs). Features include paper search, vector-based storage and retrieval, question answering, figure/table explanation, summarization, and future research direction generation. Provides an interactive Streamlit-based user interface for browsing and interacting with the system.
A multi-agent system that generates short stories through a collaborative writing process. Two specialized assistant agents take turns adding new paragraphs to the evolving narrative, building on each other's contributions. Uses Streamlit to create a user-friendly interface where humans can provide initial prompts and view the generated story.
The goal of this project is to showcase the capabilities of LLM-powered multi-agent systems in tackling complex, collaborative tasks like academic research assistance and creative writing.