TY - GEN
T1 - Topic-Oriented Open Relation Extraction with A Priori Seed Generation
AU - Ding, Linyi
AU - Xiao, Jinfeng
AU - Zhou, Sizhe
AU - Yang, Chaoqi
AU - Han, Jiawei
N1 - Research was supported in part by US DARPA INCAS Program No. HR0011-21-C0165 and BRIES Program No. HR0011-24-3-0325, National Science Foundation IIS-19-56151, the Molecule Maker Lab Institute: An AI Research Institutes program supported by NSF under Award No. 2019897, the Institute for Geospatial Understanding through an Integrative Discovery Environment (I-GUIDE) by NSF under Award No. 2118329, and the IBM-Illinois Discovery Accelerator Institute (IIDAI). Any opinions, findings, and conclusions or recommendations expressed herein are those of the authors and do not necessarily represent the views, either expressed or implied, of IBM, DARPA, or the U.S. Government.
PY - 2024
Y1 - 2024
N2 - The field of open relation extraction (ORE) has recently observed significant advancement thanks to the growing capability of large language models (LLMs). Nevertheless, challenges persist when ORE is performed on specific topics. Existing methods give suboptimal results in five dimensions: factualness, topic relevance, informativeness, coverage, and uniformity. To improve topic-oriented ORE, we propose a zero-shot approach called PriORE: Open Relation Extraction with a Priori seed generation. PriORE leverages the built-in knowledge of LLMs to maintain a dynamic seed relation dictionary for the topic. The dictionary is initialized by seed relations generated from topic-relevant entity types and expanded during contextualized ORE. PriORE then reduces the randomness in generative ORE by converting it to a more robust relation classification task. Experiments show the approach empowers better topic-oriented control over the generated relations and thus improves ORE performance along the five dimensions, especially on specialized and narrow topics.
AB - The field of open relation extraction (ORE) has recently observed significant advancement thanks to the growing capability of large language models (LLMs). Nevertheless, challenges persist when ORE is performed on specific topics. Existing methods give suboptimal results in five dimensions: factualness, topic relevance, informativeness, coverage, and uniformity. To improve topic-oriented ORE, we propose a zero-shot approach called PriORE: Open Relation Extraction with a Priori seed generation. PriORE leverages the built-in knowledge of LLMs to maintain a dynamic seed relation dictionary for the topic. The dictionary is initialized by seed relations generated from topic-relevant entity types and expanded during contextualized ORE. PriORE then reduces the randomness in generative ORE by converting it to a more robust relation classification task. Experiments show the approach empowers better topic-oriented control over the generated relations and thus improves ORE performance along the five dimensions, especially on specialized and narrow topics.
UR - http://www.scopus.com/inward/record.url?scp=85217794869&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85217794869&partnerID=8YFLogxK
U2 - 10.18653/v1/2024.emnlp-main.766
DO - 10.18653/v1/2024.emnlp-main.766
M3 - Conference contribution
AN - SCOPUS:85217794869
T3 - EMNLP 2024 - 2024 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
SP - 13834
EP - 13845
BT - EMNLP 2024 - 2024 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
A2 - Al-Onaizan, Yaser
A2 - Bansal, Mohit
A2 - Chen, Yun-Nung
PB - Association for Computational Linguistics (ACL)
T2 - 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024
Y2 - 12 November 2024 through 16 November 2024
ER -