NTR Webinar: Unsupervised Quality Estimation for Neural Machine Translation

NTR organizes and hosts scientific webinars on neural networks and invites speakers from all over the world to present their recent work at the webinars.

On May 25 Marina Fomicheva, University of Sheffield, Sheffield, UK, led a technical Zoom webinar on Unsupervised Quality Estimation for Neural Machine Translation

About the webinar: 

Most of the current state-of-the-art approaches to automatic evaluation of Natural Language Generation (NLG) systems treat the evaluated model as a black box. 

Ms. Fomicheva presented an alternative approach that looks inside the model to gain insights on the quality of generated outputs, using neural Machine Translation (MT) as an example. 

The proposed approach explores recent advances in uncertainty quantification for neural networks and performs competitively while removing the need for training heavy dedicated models for quality estimation.

Webinar recording

Webinar presentation

Moderator and contact: 

NTR CEO Nick Mikhailovsky: nickm@ntrlab.com.

Leave a Reply

Your email address will not be published. Required fields are marked *