Verb Physics: Relative Physical Knowledge of Actions and Objects

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 266–276 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1025 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 266–276 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1025 VERB PHYSICS: Relative Physical Knowledge of Actions and Objects Maxwell Forbes Yejin Choi Paul G. Allen School of Computer Science & Engineering University of Washington {mbforbes,yejin}@cs.washington.edu Abstract Learning commonsense knowledge from natural language text is nontrivial due to reporting bias: people rarely state the ob- vious, e.g., “My house is bigger than me.” However, while rarely stated explicitly, this trivial everyday knowledge does influ- ence the way people talk about the world, which provides indirect clues to reason about the world. For example, a statement like, “Tyler entered his house” implies that his house is bigger than Tyler. In this paper, we present an approach to infer relative physical knowledge of ac- tions and objects along five dimensions (e.g., size, weight, and strength) from un- structured natural language text. We frame knowledge acquisition as joint inference over two closely related problems: learn- ing (1) relative physical knowledge of ob- ject pairs and (2) physical implications of actions when applied to those object pairs. Empirical results demonstrate that it is possible to extract knowledge of actions and objects from language and that joint inference over different types of knowl- edge improves performance. 1 Introduction Reading and reasoning about natural language text often requires trivial knowledge about everyday physical actions and objects. For example, given a sentence “Shanice could fit the trophy into the suitcase,” we can trivially infer that the trophy must be smaller than the suitcase even though it is not stated explicitly. This reasoning requires knowledge about the action “fit”—in particular, typical preconditions that need to be satisfied in order to perform the action. In addition, reasoning Natural language clues approx max width Relative physical knowledge about objects Physical implications of actions “She barged into the stable.” HUMAN STABLE size: smaller weight: lighter speed: faster strength: n/a rigidness: less rigid x barged into y ⇒ x is smaller than y ⇒ x is lighter than y ⇒ x is faster than y ⇒ x is less rigid than y Figure 1: An overview of our approach. A verb’s usage in language (top) implies physical relations between objects it takes as arguments. This allows us to reason about properties of specific objects (middle), as well as the knowledge implied by the verb itself (bottom). about the applicability of various physical actions in a given situation often requires background knowledge about objects in the world, for exam- ple, that people are usually smaller than houses, that cars generally move faster than humans walk, or that a brick probably is heavier than a feather. In fact, the potential use of such knowledge about everyday actions and objects can go beyond language understanding and reasoning. Many open …

Content truncated to 500 words