About This Show
Discussing recent and interesting work related to natural language processing. Matt Gardner and Waleed Ammar, research scientists at the Allen Institute for Artificial Intelligence, give short discussions of papers, mostly in interviews with authors about their work.Read more »
Most Recent Episode
59 - Weakly Supervised Semantic Parsing With Abstract Examples, with Omer Goldman
6 days ago
ACL 2018 paper by Omer Goldman, Veronica Latcinnik, Udi Naveh, Amir Globerson, and Jonathan Berant
Omer comes on to tell us about a class project (done mostly by undergraduates!) that made it into ACL. Omer and colleagues built a semantic parser that gets state-of-the-art results on the Cornell Natural Language Visual Reasoning dataset. They did this by using "abstract examples" - they replaced the entities in the questions and corresponding logical forms with their types, labeled about a hundred examples in this abstracted formalism, and used those labels to do data augmentation and train their parser. They also used some interesting caching tricks, and a discriminative reranker.
https://www.semanticscholar.org/paper/Weakly-supervised-Semantic-Parsing-with-Abstract-Goldman-Latcinnik/5aec2ab5bf2979da067e2aa34762b589a0680030Read more »