EDBT 2025 Tutorial
基于大语言模型与知识图谱的问答技术研究:
最新进展与机遇
10:30 AM - 12:30 PM, 27th March
https://edbticdt2025.upc.edu/?contents=accepted-papers-tutorials.html
大语言模型(
LLM
)因其在自然语言理解和生成方面的卓越能力,
已在
多个问答任务中表现出
优秀的
性能。另一方面,由于
LLM
存在推理能力
弱
、领域知识过时或
匮乏
、重新训练成本高昂
、
上下文长度有限
等
问题,基于
LLM
的问答方法在
诸如
多跳问答
与
长上下文问答等复杂问答任务中
的效果不尽如人意
。
而
知识图谱(
KG
)存储
了对于推理与解释十分有效的基于图的结构化知识,可有效表达现实世界中显式的事实知识及特定领域知识
。为解决基于
LLM
的
问答方法
所面临的挑战和局限,
近期多项融合
L
LM
与
K
G
的问答研究工作被提出。本次
Tutorial
旨在概述基于
L
LM
与
K
G
的问答技术研究最新进展,并根据
K
G
的角色将相关研究工作分为三类:(
1
)
KG
作为背景知识;(
2
)
KG
作为推理
指导
;(
3
)
KG
作为
过滤器
和
验证器
。
此外,还将介绍相应
评估指标
与
基准数据集,展示特定领域行业应用
,
总结开放性挑战
及针对
数据管理
研究
的机遇。
Chuangtao Ma is a postdoctoral researcher at Aalborg University, Denmark. His research focuses on knowledge graphs, knowledge-augmented models, and their applications in data management. He is a member of the management committee of the COST action on the Global Network on Large-Scale, Cross-domain, and Multilingual Open Knowledge Graphs.
Yongrui Chen is a postdoctoral researcher at Southeast University, China. He specializes in incorporating structured and semi-structured knowledge into foundational LLMs, to improve their complex knowledge reasoning capability. He has presented numerous papers at prominent venues, including NeurIPS, TKDE, IJCAI, AAAI, ACL, ISWC, and NAACL.
Tianxing Wu is an associate professor at Southeast University, China. He is one of the main contributors to build uncertain knowledge graph reasoning tool: unKR, and schema knowledge graph: Linked Open Schema. He has published over 60 papers in top-tier venues, e.g., SIGMOD, ICDE, SIGIR, ACL, AAAI, IJCAI, and TKDE.
Arijit Khan is an IEEE senior member, an ACM distinguished speaker, and an associate professor at Aalborg University, Denmark. He published over 90 papers in premier data management and mining venues including ACM SIGMOD, VLDB, TKDE, ICDE, ICLR, USENIX ATC, and WWW. Arijit co-presented tutorials on emerging graph queries, applications, big graph systems, and graph machine learning at VLDB, ICDE, and CIKM.
Haofen Wang is a a Distinguished Researcher at Tongji University, China. He is one of the initiators of OpenKG, the world’s largest alliance for Chinese open knowledge graphs. He published over 100 highlevel papers in the AI field, and developed the world’s first interactive virtual idol–“Amber Xuyan”. Additionally, the intelligent customer service robots he built have served over 1 billion users.
OpenKG
OpenKG(中文开放知识图谱)旨在推动以中文为核心的知识图谱数据的开放、互联及众包,并促进知识图谱算法、工具及平台的开源开放。
点击
阅读原文
,进入 OpenKG 网站。