Large-scale multi-document summarization dataset and code

SUPERT: Unsupervised multi-document summarization evaluation & generation

A simple python implementation of the Maximal Marginal Relevance (MMR) baseline system for text summarization.

Multi-XScience: A Large-scale Dataset for Extreme Multi-document Summarization of Scientific Articles

Code for "Multi-document Summarization with Maximal Marginal Relevance-guided Reinforcement Learning", EMNLP 2020

Code for the paper: Improving Multi-Document Summarization through Referenced Flexible Extraction with Credit-Awareness

Extractive Multi-document Summarization

A multilingual and multi-document model that uses an enhanced version of TF-IDF and knowledge graphs to generate an abstractive summary

LongT5-based model pre-trained on a large amount of unlabeled Vietnamese news texts and fine-tuned with ViMS and VMDS collections

[Computer Speech & Language, Elsevier] - Neural Sentence Fusion for Diversity Driven Abstractive Multi-Document Summarization.

An Automatic Answer Summariser developed using Python, PyTorch, and HuggingFace trained on Quora Dataset aimed at summarizing and providing a single answer to a question using answers from multiple users.

This project uses transfer learning with the T5 model to summarize multiple documents into concise summaries. It leverages advanced NLP techniques to efficiently process and extract key information from large text datasets.

Abstractive multi-document summarization pipeline over the Multi-News dataset using three state-of-the-art seq2seq models: Pegasus-Large, BART-Large-CNN, and LED-Base-16384

Thesis project for the MSc "Language Technology" of the National and Kapodistrian University of Athens (NKUA)

Hybrid multi-document summarization system using Sentence-Transformers and KMeans clustering for semantic topic extraction, followed by BART-based abstractive refinement. Fully containerized with FastAPI and Docker.