notesum.ai

Published at December 6

A*Net and NBFNet Learn Negative Patterns on Knowledge Graphs

cs.AI

Released Date: December 6, 2024

Authors: Patrick Betz¹, Nathanael Stelzner¹, Christian Meilicke¹, Heiner Stuckenschmidt¹, Christian Bartelt¹

Aff.: ¹University of Mannheim

Arxiv: http://arxiv.org/pdf/2412.05114v1

	Approach	MRR Tail	MRR Head	MRR
	Rules	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00
	ComplEx	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00
ROH off	NBFNet	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00
ROH off	A*Net	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00	1.00 $\pm$ 0.00
ROH on	NBFNet	0.22 $\pm$ 0.04	0.33 $\pm$ 0.00	0.28 $\pm$ 0.02
ROH on	A*Net	0.18 $\pm$ 0.30	0.22 $\pm$ 0.13	0.20 $\pm$ 0.19