Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

A Repository of Conversational Datasets

About

Progress in Machine Learning is often driven by the availability of large datasets, and consistent evaluation metrics for comparing modeling approaches. To this end, we present a repository of conversational datasets consisting of hundreds of millions of examples, and a standardised evaluation procedure for conversational response selection models using '1-of-100 accuracy'. The repository contains scripts that allow researchers to reproduce the standard datasets, or to adapt the pre-processing and data filtering steps to their needs. We introduce and evaluate several competitive baselines for conversational response selection, whose implementations are shared in the repository, as well as a neural encoder model that is trained on the entire training set.

Matthew Henderson, Pawe{\l} Budzianowski, I\~nigo Casanueva, Sam Coope, Daniela Gerz, Girish Kumar, Nikola Mrk\v{s}i\'c, Georgios Spithourakis, Pei-Hao Su, Ivan Vuli\'c, Tsung-Hsien Wen• 2019

Related benchmarks

TaskDatasetResultRank
Response SelectionReddit (test)
R@1 (R100)47.7
7
Response SelectionAmazonQA (test)
R@1 (K=100)70.7
6
Showing 2 of 2 rows

Other info

Code

Follow for update