Edge-centric optimization: a novel strategy for minimizing information loss in graph-to-text generation

doi:10.1007/s40747-024-01690-y

	Edge-centric optimization: a novel strategy for minimizing information loss in graph-to-text generation
	Yao, Zheng 2; Li, Jingyuan 1; Cen, Jianhe 2; Sun, Shiqi 2; Yin, Dahu 2; Wang, Yuanzhuo 3
	2025
发表期刊	COMPLEX & INTELLIGENT SYSTEMS
ISSN	2199-4536
卷号	11 期号:1 页码:19
摘要	Given the remarkable text generation capabilities of pre-trained language models, impressive results have been realized in graph-to-text generation. However, while learning from knowledge graphs, these language models are unable to fully grasp the structural information of the graph, leading to logical errors and missing key information. Therefore, an important research direction is to minimize the loss of graph structural information during the model training process. We propose a framework named Edge-Optimized Multi-Level Information refinement (EMLR), which aims to maximize the retention of the graph's structural information from an edge perspective. Based on this framework, we further propose a new graph generation model, named TriELMR, highlighting the comprehensive interactive learning relationship between the model and the graph structure, as well as the importance of edges in the graph structure. TriELMR adopts three main strategies to reduce information loss during learning: (1) Knowledge Sequence Optimization; (2) EMLR Framework; and (3) Graph Activation Function. Experimental results reveal that TriELMR exhibits exceptional performance across various benchmark tests, especially on the webnlgv2.0 and Event Narrative datasets, achieving BLEU-4 scores of 66.5% and 37.27%, respectively, surpassing the state-of-the-art models. These demonstrate the advantages of TriELMR in maintaining the accuracy of graph structural information.
关键词	Graph-generated text Knowledge representation Knowledge graphs Graph theory Edge-aware attention
DOI	10.1007/s40747-024-01690-y
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[62172393] ; Henan Province Key Research and Development Project[241111211900] ; Zhongyuanyingcai program[204200510002] ; Ministry of Education Industry Education Collaborative Education Project by Tencent[230700006203144]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:001380505300001
出版者	SPRINGER HEIDELBERG
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41067
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Li, Jingyuan; Wang, Yuanzhuo
作者单位	1.Beijing Technol & Business Univ, Sch Comp & Artificial Intelligence, Beijing 100048, Peoples R China 2.Zhengzhou Univ, Henan Inst Adv Technol, Zhengzhou 450001, Henan, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Yao, Zheng,Li, Jingyuan,Cen, Jianhe,et al. Edge-centric optimization: a novel strategy for minimizing information loss in graph-to-text generation[J]. COMPLEX & INTELLIGENT SYSTEMS,2025,11(1):19.
APA	Yao, Zheng,Li, Jingyuan,Cen, Jianhe,Sun, Shiqi,Yin, Dahu,&Wang, Yuanzhuo.(2025).Edge-centric optimization: a novel strategy for minimizing information loss in graph-to-text generation.COMPLEX & INTELLIGENT SYSTEMS,11(1),19.
MLA	Yao, Zheng,et al."Edge-centric optimization: a novel strategy for minimizing information loss in graph-to-text generation".COMPLEX & INTELLIGENT SYSTEMS 11.1(2025):19.