PocketSkye's picture
Update README.md
2d24d2d verified
|
Raw
History Blame Contribute Delete
1.64 kB
---
title: MultiAgent Research Assistant
emoji: 💬
colorFrom: yellow
colorTo: purple
sdk: gradio
sdk_version: 5.0.1
app_file: app.py
pinned: true
license: apache-2.0
short_description: A MultiAgent research assistant and critique
tags:
- agent-demo-track
---
Multi-Agent Research Assistant
A powerful research assistant built using LangChain agents, LangGraph, and Hugging Face Transformers, designed to automate the research process across academic topics. This app retrieves papers, summarizes them, critiques findings, and generates a final synthesized report—all via coordinated AI agents.
🚀 Demo (Hugging Face Space)
Enter your research topic and let the assistant do the heavy lifting.
🧩 Agents Overview
Agent Description
Retriever Queries ArXiv, Semantic Scholar, OpenAlex for relevant academic papers.
Summarizer Extracts core ideas and results from each paper using Mistral-small.
Critique Compares papers, detects contradictions, and flags hype/missing data.
Synthesizer Generates a final report with structured summary and analysis.
🛠️ Tech Stack
LangChain for agent orchestration
Mistral models (Tiny, Small, Medium, Large) via Hugging Face Inference APIs
FAISS for optional RAG (Retrieval-Augmented Generation)
PDFPlumber for extracting full-text content
Research APIs: ArXiv, Semantic Scholar, OpenAlex, PapersWithCode
Gradio UI for simple user interaction
An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).