Multidirectional Associative Optimization of Function-Specific Word Representations
This addresses the challenge of reasoning over SVO structures in natural language processing, offering a versatile and efficient solution for tasks like selectional preference and event similarity.
The paper tackles the problem of learning associations between interrelated word groups like Subject-Verb-Object structures by inducing a joint function-specific word vector space, resulting in state-of-the-art performance on selectional preference and event similarity tasks while reducing parameters by up to 95%.
We present a neural framework for learning associations between interrelated groups of words such as the ones found in Subject-Verb-Object (SVO) structures. Our model induces a joint function-specific word vector space, where vectors of e.g. plausible SVO compositions lie close together. The model retains information about word group membership even in the joint space, and can thereby effectively be applied to a number of tasks reasoning over the SVO structure. We show the robustness and versatility of the proposed framework by reporting state-of-the-art results on the tasks of estimating selectional preference and event similarity. The results indicate that the combinations of representations learned with our task-independent model outperform task-specific architectures from prior work, while reducing the number of parameters by up to 95%.