A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation
This work addresses the challenge of commonsense story generation for natural language processing applications, representing an incremental improvement by integrating external knowledge and multi-task learning.
The paper tackled the problem of generating coherent and logical stories by addressing issues like repetition and lack of long-range coherence in existing models, resulting in a knowledge-enhanced pretraining model that outperforms state-of-the-art baselines in logic and global coherence.
Story generation, namely generating a reasonable story from a leading context, is an important but challenging task. In spite of the success in modeling fluency and local coherence, existing neural language generation models (e.g., GPT-2) still suffer from repetition, logic conflicts, and lack of long-range coherence in generated stories. We conjecture that this is because of the difficulty of associating relevant commonsense knowledge, understanding the causal relationships, and planning entities and events with proper temporal order. In this paper, we devise a knowledge-enhanced pretraining model for commonsense story generation. We propose to utilize commonsense knowledge from external knowledge bases to generate reasonable stories. To further capture the causal and temporal dependencies between the sentences in a reasonable story, we employ multi-task learning which combines a discriminative objective to distinguish true and fake stories during fine-tuning. Automatic and manual evaluation shows that our model can generate more reasonable stories than state-of-the-art baselines, particularly in terms of logic and global coherence.