https://github.com/ovuiproduction/research-assistant
AI-Powered Research Assistant – A smart tool that helps researchers find relevant papers, recommend journals, ask questions about content, humanize AI text, and detect AI-generated writing. Powered by Large Language Models for enhanced research productivity.
https://github.com/ovuiproduction/research-assistant
all-minilm-l6-v2 artificial-intelligence bart faiss fine-tuning humanizer natural-language-processing question-answering rag research-assistant search-engine text-processing
Last synced: 3 months ago
JSON representation
AI-Powered Research Assistant – A smart tool that helps researchers find relevant papers, recommend journals, ask questions about content, humanize AI text, and detect AI-generated writing. Powered by Large Language Models for enhanced research productivity.
- Host: GitHub
- URL: https://github.com/ovuiproduction/research-assistant
- Owner: ovuiproduction
- License: mit
- Created: 2025-05-16T07:37:21.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-08-06T19:45:56.000Z (12 months ago)
- Last Synced: 2025-08-06T21:34:37.766Z (12 months ago)
- Topics: all-minilm-l6-v2, artificial-intelligence, bart, faiss, fine-tuning, humanizer, natural-language-processing, question-answering, rag, research-assistant, search-engine, text-processing
- Language: JavaScript
- Homepage:
- Size: 1.16 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🎓 AI-ResearchMate: Humanizing and Assisting Research with LLMs
### [Demo](#)
An AI-powered co-research assistant that simplifies academic research using intelligent search, interactive Q&A, and humanized text transformation to help bypass AI content detection mechanisms.
---
## 🚀 Objective
**AI-ResearchMate** is designed to:
- Assist researchers in finding high-quality academic content.
- Answer research questions using a RAG (Retrieval-Augmented Generation) pipeline.
- Humanize AI-generated content to avoid AI detection.
- Identify and highlight AI-written sections in research drafts and rewrite them in a more human tone.
---
## 🧠 Features
- 🔍 **Smart Paper Search**: Retrieve top relevant research papers to basis of abstract , title or .author names.
- 🔍 **Smart Journal Search**: Suggesting Journals based on domain and keywords also filterring Qurtile based.
- 💬 **Interactive Q&A**: Ask questions about research content and get answers powered by Large Language Models.
- 📝 **AI Content Humanization**: Use BART-based pipeline to rewrite AI-generated text into human-like writing.
- 🚨 **AI Content Detection**: Detect and highlight AI-written content to improve authenticity and originality.
---
## 🧰 Tech Stack
- **Language**: Python
- **LLM Models**: BART (`facebook/bart-large` via HuggingFace) , all-MiniLM-L6-v2
- **Vector Search**: FAISS
- **Database**: MongoDB
- **Datasets**:
- ArXiv Paper Dataset
- Journal Ranking Dataset
- Self-Curated AI-Human Text Pair Dataset:
---
## 🛠️ Key Contributions
- 🔧 Developed a RAG-based **research & journal retrieval system** using document embeddings stored in FAISS.
- ✍️ Engineered a **BART-powered content humanization pipeline** that rewrites detected AI content to human tone.
- 🧪 Created an **AI content detector** to flag AI-written sections and replace them with more natural writing.
---
## 📦 Setup Instructions
1. **Clone the Repository**
```bash
git clone https://github.com/yourusername/AI-ResearchMate.git
cd AI-ResearchMate
```
## License
This project is licensed under the [MIT License](https://github.com/ovuiproduction/Research-Assistant/blob/main/LICENSE)