Comprehension Datasets

Reading comprehension datasets are crucial for training and evaluating question-answering systems, particularly in low-resource languages and specialized domains. Current research focuses on improving dataset quality through techniques like data augmentation using large language models (LLMs) to generate synthetic data, thereby addressing issues of size and robustness. This work aims to create more comprehensive and representative datasets, leading to more accurate and reliable question-answering systems with improved performance across diverse contexts and distribution shifts. The resulting advancements have significant implications for various applications, including information retrieval, knowledge extraction, and personalized learning.

Papers

September 27, 2023

Question answering using deep learning in low resource Indian language Marathi
Dhiraj Amin, Sharvari Govilkar, Sagar Kulkarni
Deep Learning Yes No Question Low Resource Bidirectional Encoder Representation Marathi Corpus Comprehension Datasets

September 21, 2023

Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges
Vinay Samuel, Houda Aynaou, Arijit Ghosh Chowdhury, Karthik Venkat Ramanan, Aman Chadha
Technical Challenge Emerging Opportunity NLP Task Reading Comprehension Comprehension Datasets

September 3, 2023

Generative Data Augmentation using LLMs improves Distributional Robustness in Question Answering
Arijit Ghosh Chowdhury, Aman Chadha
Natural Language Processing Generative Model Domain Adaptation Medical LLM Question Answering Distributional Robustness Generative Data Augmentation Comprehension Datasets

May 23, 2023

NarrativeXL: A Large-scale Dataset For Long-Term Memory Models
Arseny Moskvichev, Ky-Vinh Mai
Language Model Long Short Term Memory Large Scale Dataset Context Length Modern Language Model Scene Recognition Comprehension Datasets

April 2, 2023

A Data-centric Framework for Improving Domain-specific Machine Reading Comprehension Datasets
Iva Bojic, Josef Halim, Verena Suharman, Sreeja Tar, Qi Chwen Ong, Duy Phung, Mathieu Ravaut, Shafiq Joty, Josip Car
High Quality Data Centric BioASQ Challenge Dataset Quality Comprehension Datasets

Comprehension Datasets

Papers

Question answering using deep learning in low resource Indian language Marathi

Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges

Generative Data Augmentation using LLMs improves Distributional Robustness in Question Answering

NarrativeXL: A Large-scale Dataset For Long-Term Memory Models

A Data-centric Framework for Improving Domain-specific Machine Reading Comprehension Datasets