Graph-based Multi-hop Reasoning for Long Text Generation
This addresses the problem of generating coherent long text for applications like story and review generation, though it appears incremental as it builds on existing knowledge-enhanced models.
The paper tackles the challenge of learning sentence-level semantic dependencies in long text generation by proposing a Multi-hop Reasoning Generation (MRG) approach that uses multi-hop reasoning over a knowledge graph, resulting in more informative and coherent text compared to strong baselines like GPT-2.
Long text generation is an important but challenging task.The main problem lies in learning sentence-level semantic dependencies which traditional generative models often suffer from. To address this problem, we propose a Multi-hop Reasoning Generation (MRG) approach that incorporates multi-hop reasoning over a knowledge graph to learn semantic dependencies among sentences. MRG consists of twoparts, a graph-based multi-hop reasoning module and a path-aware sentence realization module. The reasoning module is responsible for searching skeleton paths from a knowledge graph to imitate the imagination process in the human writing for semantic transfer. Based on the inferred paths, the sentence realization module then generates a complete sentence. Unlike previous black-box models, MRG explicitly infers the skeleton path, which provides explanatory views tounderstand how the proposed model works. We conduct experiments on three representative tasks, including story generation, review generation, and product description generation. Automatic and manual evaluation show that our proposed method can generate more informative and coherentlong text than strong baselines, such as pre-trained models(e.g. GPT-2) and knowledge-enhanced models.