CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning

TL;DR

CommonGen introduces a new constrained text generation task and dataset to evaluate and improve generative commonsense reasoning in AI models, highlighting current gaps and potential for transfer learning.

Contribution

The paper presents a novel benchmark dataset and task for generative commonsense reasoning, emphasizing relational reasoning and compositional generalization.

Findings

Large gap between models and human performance

Transfer learning improves downstream commonsense tasks

Dataset contains 79k descriptions over 35k concept-sets

Abstract

Recently, large-scale pre-trained language models have demonstrated impressive performance on several commonsense-reasoning benchmark datasets. However, building machines with commonsense to compose realistically plausible sentences remains challenging. In this paper, we present a constrained text generation task, CommonGen associated with a benchmark dataset, to explicitly test machines for the ability of generative commonsense reasoning. Given a set of common concepts (e.g., {dog, frisbee, catch, throw}); the task is to generate a coherent sentence describing an everyday scenario using these concepts (e.g., "a man throws a frisbee and his dog catches it"). The CommonGen task is challenging because it inherently requires 1) relational reasoning with background commonsense knowledge, and 2) compositional generalization ability to work on unseen concept combinations. Our dataset, constructed through a combination of crowdsourced and existing caption corpora, consists of 79k commonsense descriptions over 35k unique concept-sets. Experiments show that there is a large gap between state-of-the-art text generation models (e.g., T5) and human performance. Furthermore, we demonstrate that the learned generative commonsense reasoning capability can be transferred to improve downstream tasks such as CommonsenseQA by generating additional context.