Is Architectural Complexity Overrated? Competitive and Interpretable Knowledge Graph Completion with RelatE

TL;DR

RelatE is a simple, interpretable, real-valued embedding model for knowledge graph completion that outperforms complex models in accuracy, efficiency, and robustness, while maintaining full expressiveness.

Contribution

The paper introduces RelatE, a novel real-valued, interpretable embedding model that efficiently encodes relational patterns and surpasses prior complex models in performance and robustness.

Findings

RelatE achieves higher MRR and Hit@10 on YAGO3-10 than all baselines.

RelatE reduces training time, inference latency, and GPU memory usage significantly.

Perturbation studies show improved robustness against structural edits.

Abstract

We revisit the efficacy of simple, real-valued embedding models for knowledge graph completion and introduce RelatE, an interpretable and modular method that efficiently integrates dual representations for entities and relations. RelatE employs a real-valued phase-modulus decomposition, leveraging sinusoidal phase alignments to encode relational patterns such as symmetry, inversion, and composition. In contrast to recent approaches based on complex-valued embeddings or deep neural architectures, RelatE preserves architectural simplicity while achieving competitive or superior performance on standard benchmarks. Empirically, RelatE outperforms prior methods across several datasets: on YAGO3-10, it achieves an MRR of 0.521 and Hit@10 of 0.680, surpassing all baselines. Additionally, RelatE offers significant efficiency gains, reducing training time by 24%, inference latency by 31%, and peak GPU memory usage by 22% compared to RotatE. Perturbation studies demonstrate improved robustness, with MRR degradation reduced by up to 61% relative to TransE and by up to 19% compared to RotatE under structural edits such as edge removals and relation swaps. Formal analysis further establishes the model's full expressiveness and its capacity to represent essential first-order logical inference patterns. These results position RelatE as a scalable and interpretable alternative to more complex architectures for knowledge graph completion.