Papers › Augmenting Compositional Models for Knowledge Base Completion Using Gradient Representations

Augmenting Compositional Models for Knowledge Base Completion Using Gradient Representations

2 Nov 2018arXiv:1811.01062archive 2025-07-28

Matthias Lalisse, Paul Smolensky

Neural models of Knowledge Base data have typically employed compositional representations of graph objects: entity and relation embeddings are systematically combined to evaluate the truth of a candidate Knowedge Base entry. Using a model inspired by Harmonic Grammar, we propose to tokenize triplet embeddings by subjecting them to a process of optimization with respect to learned well-formedness conditions on Knowledge Base triplets. The resulting model, known as Gradient Graphs, leads to sizable improvements when implemented as a companion to compositional models. Also, we show that the "supracompositional" triplet token embeddings it produces have interpretable properties that prove helpful in performing inference on the resulting triplet representations.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

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

Tasks

Knowledge Base CompletionKnowledge GraphsLink Prediction

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Knowledge Graphs FB15k HHolE MRR .796 #1 of 2 Archive leaderboard report
Link Prediction FB15k HHolE Hits@1 0.727 #4 of 10 Archive leaderboard report
Link Prediction FB15k HHolE Hits@10 0.901 #4 of 10 Archive leaderboard report
Link Prediction FB15k HHolE Hits@3 0.848 #4 of 10 Archive leaderboard report
Link Prediction FB15k HHolE MR 21 #4 of 10 Archive leaderboard report
Link Prediction FB15k HHolE MRR 0.796 #4 of 10 Archive leaderboard report
Link Prediction FB15k HHolE Hits@1 .727 #8 of 23 Archive leaderboard report
Link Prediction FB15k HHolE Hits@10 .901 #8 of 23 Archive leaderboard report
Link Prediction FB15k HHolE Hits@3 .848 #8 of 23 Archive leaderboard report
Link Prediction FB15k HHolE MR 21 #8 of 23 Archive leaderboard report
Link Prediction FB15k HHolE MRR .796 #8 of 23 Archive leaderboard report
Link Prediction WN18 HHolE Hits@1 .931 #20 of 37 Archive leaderboard report
Link Prediction WN18 HHolE Hits@10 .951 #20 of 37 Archive leaderboard report
Link Prediction WN18 HHolE Hits@3 .945 #20 of 37 Archive leaderboard report
Link Prediction WN18 HHolE MR 183 #20 of 37 Archive leaderboard report
Link Prediction WN18 HHolE MRR .939 #20 of 37 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