Is Structure Necessary for Modeling Argument Expectations in Distributional Semantics?
This work addresses a fundamental question in NLP about event representation for argument prediction, but it is incremental as it builds on existing models and experimental evidence.
The paper investigates whether structured event representations are necessary for modeling verb argument expectations, comparing structured and unstructured models on the task of updating argument expectations.
Despite the number of NLP studies dedicated to thematic fit estimation, little attention has been paid to the related task of composing and updating verb argument expectations. The few exceptions have mostly modeled this phenomenon with structured distributional models, implicitly assuming a similarly structured representation of events. Recent experimental evidence, however, suggests that human processing system could also exploit an unstructured "bag-of-arguments" type of event representation to predict upcoming input. In this paper, we re-implement a traditional structured model and adapt it to compare the different hypotheses concerning the degree of structure in our event knowledge, evaluating their relative performance in the task of the argument expectations update.