Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model
The task of multi-hop question generation (QG) seeks to generate questions that require a complex reasoning process that spans multiple sentences and answers. Beyond the conventional challenges of what to ask and how to ask, multi-hop QG necessitates sophisticated reasoning from dispersed evidence a...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/9/5765 |
| _version_ | 1797602946820603904 |
|---|---|
| author | Zhenping Li Zhen Cao Pengfei Li Yong Zhong Shaobo Li |
| author_facet | Zhenping Li Zhen Cao Pengfei Li Yong Zhong Shaobo Li |
| author_sort | Zhenping Li |
| collection | DOAJ |
| description | The task of multi-hop question generation (QG) seeks to generate questions that require a complex reasoning process that spans multiple sentences and answers. Beyond the conventional challenges of what to ask and how to ask, multi-hop QG necessitates sophisticated reasoning from dispersed evidence across multiple sentences. To address these challenges, a knowledge graph-enhanced language model (KGEL) has been developed to imitate human reasoning for multi-hop questions.The initial step in KGEL involves encoding the input sentence with a pre-trained GPT-2 language model to obtain a comprehensive semantic context representation. Next, a knowledge graph is constructed using the entities identified within the context. The critical information in the graph that is related to the answer is then utilized to update the context representations through an answer-aware graph attention network (GAT). Finally, the multi-head attention generation module (MHAG) is performed over the updated latent representations of the context to generate coherent questions. Human evaluations demonstrate that KGEL generates more logical and fluent multi-hop questions compared to GPT-2. Furthermore, KGEL outperforms five prominent baselines in automatic evaluations, with a BLEU-4 score that is 27% higher than that of GPT-2. |
| first_indexed | 2024-03-11T04:23:43Z |
| format | Article |
| id | doaj.art-eff6ab4e02114366bbf33b0deb4c2165 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T04:23:43Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-eff6ab4e02114366bbf33b0deb4c21652023-11-17T22:38:30ZengMDPI AGApplied Sciences2076-34172023-05-01139576510.3390/app13095765Multi-Hop Question Generation with Knowledge Graph-Enhanced Language ModelZhenping Li0Zhen Cao1Pengfei Li2Yong Zhong3Shaobo Li4Chengdu Institute of Computer Applications, Chinese Academy of Sciences, Chengdu 610041, ChinaSchool of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798, SingaporeSchool of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798, SingaporeChengdu Institute of Computer Applications, Chinese Academy of Sciences, Chengdu 610041, ChinaChengdu Institute of Computer Applications, Chinese Academy of Sciences, Chengdu 610041, ChinaThe task of multi-hop question generation (QG) seeks to generate questions that require a complex reasoning process that spans multiple sentences and answers. Beyond the conventional challenges of what to ask and how to ask, multi-hop QG necessitates sophisticated reasoning from dispersed evidence across multiple sentences. To address these challenges, a knowledge graph-enhanced language model (KGEL) has been developed to imitate human reasoning for multi-hop questions.The initial step in KGEL involves encoding the input sentence with a pre-trained GPT-2 language model to obtain a comprehensive semantic context representation. Next, a knowledge graph is constructed using the entities identified within the context. The critical information in the graph that is related to the answer is then utilized to update the context representations through an answer-aware graph attention network (GAT). Finally, the multi-head attention generation module (MHAG) is performed over the updated latent representations of the context to generate coherent questions. Human evaluations demonstrate that KGEL generates more logical and fluent multi-hop questions compared to GPT-2. Furthermore, KGEL outperforms five prominent baselines in automatic evaluations, with a BLEU-4 score that is 27% higher than that of GPT-2.https://www.mdpi.com/2076-3417/13/9/5765multi-hop question generationgraph neural networknatural language processingreasoning chain |
| spellingShingle | Zhenping Li Zhen Cao Pengfei Li Yong Zhong Shaobo Li Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model Applied Sciences multi-hop question generation graph neural network natural language processing reasoning chain |
| title | Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model |
| title_full | Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model |
| title_fullStr | Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model |
| title_full_unstemmed | Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model |
| title_short | Multi-Hop Question Generation with Knowledge Graph-Enhanced Language Model |
| title_sort | multi hop question generation with knowledge graph enhanced language model |
| topic | multi-hop question generation graph neural network natural language processing reasoning chain |
| url | https://www.mdpi.com/2076-3417/13/9/5765 |
| work_keys_str_mv | AT zhenpingli multihopquestiongenerationwithknowledgegraphenhancedlanguagemodel AT zhencao multihopquestiongenerationwithknowledgegraphenhancedlanguagemodel AT pengfeili multihopquestiongenerationwithknowledgegraphenhancedlanguagemodel AT yongzhong multihopquestiongenerationwithknowledgegraphenhancedlanguagemodel AT shaoboli multihopquestiongenerationwithknowledgegraphenhancedlanguagemodel |