KGR^4: Retrieval, Retrospect, Refine and Rethink for Commonsense Generation
This addresses the challenge of improving generative commonsense reasoning for AI systems, offering a novel framework that significantly enhances performance on a specific benchmark.
The paper tackles the problem of generating plausible and grammatically correct sentences for commonsense reasoning by proposing KGR^4, a framework that retrieves prototypes, edits them, refines errors, and selects outputs, achieving a state-of-the-art SPICE score of 33.56 on the CommonGen benchmark, which is 2.49 points higher than previous best results.
Generative commonsense reasoning requires machines to generate sentences describing an everyday scenario given several concepts, which has attracted much attention recently. However, existing models cannot perform as well as humans, since sentences they produce are often implausible and grammatically incorrect. In this paper, inspired by the process of humans creating sentences, we propose a novel Knowledge-enhanced Commonsense Generation framework, termed KGR^4, consisting of four stages: Retrieval, Retrospect, Refine, Rethink. Under this framework, we first perform retrieval to search for relevant sentences from external corpus as the prototypes. Then, we train the generator that either edits or copies these prototypes to generate candidate sentences, of which potential errors will be fixed by an autoencoder-based refiner. Finally, we select the output sentence from candidate sentences produced by generators with different hyper-parameters. Experimental results and in-depth analysis on the CommonGen benchmark strongly demonstrate the effectiveness of our framework. Particularly, KGR^4 obtains 33.56 SPICE points in the official leaderboard, outperforming the previously-reported best result by 2.49 SPICE points and achieving state-of-the-art performance.