Contextual Modeling for Language and Vision with Nasrin Mostafazadeh

Today we’re joined by Nasrin Mostafazadeh, Senior AI Research Scientist at New York-based Elemental Cognition.

Subscribe: iTunes / Google Play / Spotify / RSS

Our conversation focuses on Nasrin’s work in event-centric contextual modeling in language and vision, which she sees as a means of giving AI systems a bit of “common sense.” We discuss Nasrin’s work on the Story Cloze Test, which is a reasoning framework for evaluating story understanding and generation. We explore the details of this task–including what constitutes a “story”–and some of the challenges it presents and approaches for solving it. We also discuss how you model what a computer understands, building semantic representation algorithms, different ways to approach “explainability,” and multimodal extensions to her contextual modeling work.

Sign up for our Newsletter!

Be sure to sign up for our weekly newsletter. We recently shared a write up detailing the ML/AI Job Board we’re working on, and got a ton of encouragement and interest. To make sure you don’t miss anything, head over to twimlai.com/newsletter to sign up.

About Nasrin

• Nasrin on Linkedin
• Nasrin’s Personal Page
• Nasrin on Twitter

Mentioned in the Interview

• Elemental Cognition
• Event-centric Multimodal Context Modeling
• Story Cloze Test
• Paper: Tackling the Story Ending Biases in The Story Cloze Test
• Paper: Improving Language Understanding by Generative Pre-Training
• Paper: The Effect of Different Writing Tasks on Linguistic Style: A Case Study of the ROC Story Cloze Task
• Paper: deltaBLEU: A Discriminative Metric for Generation Tasks with Intrinsically Diverse Targets
• Paper: BLEU: a Method for Automatic Evaluation of Machine Translation
• TWiML Presents: Series page
• TWiML Events Page
• TWiML Meetup
• TWiML Newsletter

“More On That Later” by Lee Rosevere licensed under CC By 4.0