Papers › Friendly Neighbors: Contextualized Sequence-to-Sequence Link Prediction

Friendly Neighbors: Contextualized Sequence-to-Sequence Link Prediction

22 May 2023arXiv:2305.13059archive 2025-07-28

Adrian Kochsiek, Apoorv Saxena, Inderjeet Nair, Rainer Gemulla

We propose KGT5-context, a simple sequence-to-sequence model for link prediction (LP) in knowledge graphs (KG). Our work expands on KGT5, a recent LP model that exploits textual features of the KG, has small model size, and is scalable. To reach good predictive performance, however, KGT5 relies on an ensemble with a knowledge graph embedding model, which itself is excessively large and costly to use. In this short paper, we show empirically that adding contextual information - i.e., information about the direct neighborhood of the query entity - alleviates the need for a separate KGE model to obtain good performance. The resulting KGT5-context model is simple, reduces model size significantly, and obtains state-of-the-art performance in our experimental study.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

uma-pi1/kgt5-context officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Graph Embedding Knowledge Graph Embedding Knowledge Graphs Link Prediction Prediction

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Link Prediction	Wikidata5M	KGT5-context + Description	Hits@1	0.406	#2 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context + Description	Hits@10	0.46	#2 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context + Description	Hits@3	0.44	#2 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context + Description	MRR	0.426	#2 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5 + Description	Hits@1	0.357	#3 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5 + Description	Hits@10	0.422	#3 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5 + Description	Hits@3	0.397	#3 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5 + Description	MRR	0.381	#3 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context	Hits@1	0.35	#4 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context	Hits@10	0.427	#4 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context	Hits@3	0.396	#4 of 14	Archive leaderboard	report
Link Prediction	Wikidata5M	KGT5-context	MRR	0.378	#4 of 14	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections