bart-model

NeuroDoc is a powerful AI-based offline document summarization tool that leverages OCR and NLP to intelligently analyze PDFs and generate structured summaries. Built using Flask, this tool is designed to run completely offline and supports both text-based and scanned/image-based documents.

This repository demonstrates how to use Hugging Face Transformers for text summarization. We focus on two state-of-the-art models: BART (facebook/bart-large-cnn) T5 (t5-large) Both models are designed for sequence-to-sequence tasks, making them ideal for text summarization.

Chrome extension for automatic webpage summarization using AI (local and API modes).

Shakespearify: A Modern English to Shakespearean Translator

Classification with semantic similarity and Summarization (LLM)

A full-stack NLP text summarization web app built with Python, Flask and HuggingFace

Information Retrieval Project : Crisis Facts Summarization

Reproducing ACL Paper for iterative text generation and performing Robustness and Multilinguality tests

API service for abstractive summarization of long texts

Pre-training BART.

Proiect Natural Language Processing (NLP), Anul 3, Semestrul 2, Facultatea de Matematica si Informatica, Universitatea din Bucuresti

Text summarization using BART model

This repository explores enhancing dialogue summarization with commonsense knowledge through the SICK framework, evaluating models on dialogue datasets to assess commonsense's impact on summarization quality.

A Software Tools & Methods Project including prompt engineering techniques for leveraging pre-trained models on Hugging Face. Concludes design, evaluation, and refining prompts for specific use cases, ensuring optimal model performance.

A Software Tools & Methods Project including prompt engineering techniques for leveraging pre-trained models on Hugging Face. Concludes design, evaluation, and refining prompts for specific use cases, ensuring optimal model performance.

An AI system that processes meetings, lectures, and documents using speech recognition, abstractive summarization, topic extraction, action-item detection, priority classification, and factual verification to generate concise, structured, and reliable insights.

A collection of text summarization projects using ensemble methods and T5 model.

This repository contains code for training and evaluating Encoder-Only/Decoder-Only/Encoder-Decoder/Enseble models using the Stanford IMDb dataset.

Complete MLOps pipeline for Abstractive Text Summarization using Hugging Face's BART-base model, fine-tuned on the scientific_papers/arxiv dataset. Features include structured training, ROUGE/BERTScore evaluation, and deployment to the Hugging Face Hub and Spaces.