Apples to Apples: Learning Semantics of Common Entities Through a Novel Comprehension Task

Abstract

Understanding common entities and their attributes is a primary requirement for any system that comprehends natural language.In order to enable learning about common entities, we introduce a novel machine comprehension task, GuessTwo: given a short paragraph comparing different aspects of two realworld semantically-similar entities, a system should guess what those entities are. Accomplishing this task requires deep language understanding which enables inference, connecting each comparison paragraph to different levels of knowledge about world entities and their attributes. So far we have crowdsourced a dataset of more than 14K comparison paragraphs comparing entities from a variety of categories such as fruits and animals. We have designed two schemes for evaluation: open-ended, and binary-choice prediction. For benchmarking further progress in the task, we have collected a set of paragraphs as the test set on which human can accomplish the task with an accuracy of 94.2% on open-ended prediction. We have implemented various models for tackling the task, ranging from semantic-driven to neural models. The semantic-driven approach outperforms the neural models, however, the results indicate that the task is very challenging across the models.

Machine Semiotics

“…This however, is not feasible at the current stage of speech and language processing technologies. Despite marvelous achievements in speech recognition technology (Böck et al 2019, Chen et al 2013, Graves et al 2013, Guo et al 2014, Mesnil et al 2015, Sundermeyer et al 2015, neither compositional semantic analysis (Allen 2014, Bakhshandeh and Allen 2017, Perera et al 2017, Socher et al 2013 nor reasonable pragmatic inferences (Benz 2016, Franke and Jäger 2016, Grice 1969, van Rooij and de Jager 2012 are currently available through artificial intelligence techniques.…”