Fine-tune open-source Language Models (LLMs) on E-commerce data, leveraging Amazon's sales data. Showcase a tailored solution for enhanced language understanding and generation with a focus on custom E-commerce datasets.

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.

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

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.

Classification with semantic similarity and Summarization (LLM)

Information Retrieval Project : Crisis Facts Summarization

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

The project contains code and resources for a sophisticated AI-driven chatbot designed to provide accurate, context-aware responses. It uses RoBerta and BART Transformers and advance NLP techniques. The chatbot is capable of handling a wide range of domains such as healthcare , finance , etc..

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.

Pre-training BART.

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

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.

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

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

Understanding Language with Transformers. Implementing and improving BERT and BART models for multiple NLP tasks like sentiment analysis, paraphrase detection, semantic similarity, and generation.