Maluuba presenting at EMNLP 2016

We’re heading to Austin, Texas next week to take part in the Conference on Empirical Methods in Natural Language Processing (EMNLP).

We’re excited to meet with fellow researchers and to take part in the workshops and tutorials at this leading conference for machine learning in NLP.

Our research team will deliver a presentation about our recent work in Natural Language Comprehension. We’ll showcase EpiReader, a novel system for machine comprehension of text. EpiReader is an end-to-end neural model comprising two components: the first compares the document with the question and proposes a small set of candidate answers; the second uses the proposed candidates and the question to formulate hypotheses, then scores the hypotheses based on their estimated entailment with the document. EpiReader achieved state-of-the-art performance on the CNN and CBT comprehension datasets.

See you there!

PS We’re hiring too! If you’d like to learn more about Maluuba come chat with us.

 

 

 

Researchpb+j Builder