Acquiring and Modelling Abstract Commonsense Knowledge via Conceptualization

TL;DR

This work tackles the insufficiency of concrete commonsense knowledge by introducing conceptualization to acquire abstract knowledge about events and situations. It proposes a framework for conceptual induction, applying Probase to ATOMIC, and building a pipeline that annotates and generates abstract knowledge, culminating in a large abstract CKG capable of inferring about unseen entities. The authors provide datasets and neural models to generate and verify abstract knowledge, and demonstrate improved performance on downstream tasks, including zero-shot commonsense QA, when augmenting CKGs with abstract knowledge. The study advances scalable abstraction in commonsense reasoning, enabling better generalization and reasoning with abstract knowledge across diverse real-world scenarios.

Abstract

Conceptualization, or viewing entities and situations as instances of abstract concepts in mind and making inferences based on that, is a vital component in human intelligence for commonsense reasoning. Despite recent progress in artificial intelligence to acquire and model commonsense attributed to neural language models and commonsense knowledge graphs (CKGs), conceptualization is yet to be introduced thoroughly, making current approaches ineffective to cover knowledge about countless diverse entities and situations in the real world. To address the problem, we thoroughly study the role of conceptualization in commonsense reasoning, and formulate a framework to replicate human conceptual induction by acquiring abstract knowledge about events regarding abstract concepts, as well as higher-level triples or inferences upon them. We then apply the framework to ATOMIC, a large-scale human-annotated CKG, aided by the taxonomy Probase. We annotate a dataset on the validity of contextualized conceptualizations from ATOMIC on both event and triple levels, develop a series of heuristic rules based on linguistic features, and train a set of neural models to generate and verify abstract knowledge. Based on these components, a pipeline to acquire abstract knowledge is built. A large abstract CKG upon ATOMIC is then induced, ready to be instantiated to infer about unseen entities or situations. Finally, we empirically show the benefits of augmenting CKGs with abstract knowledge in downstream tasks like commonsense inference and zero-shot commonsense QA.