mlsanityData to 14 Oct 2026

Recipe-Controlled Decoder Audit for Structural Knowledge-Graph Completion

arXiv 2606.14492 ↗First posted 12 Jun 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We present a recipe-controlled decoder audit (RCDA) for structural transductive knowledge-graph completion (KGC). The audit asks a simple reporting question: before attributing gains to an encoder or training recipe, what changes when the decoder is swapped under the same recipe? Using ComplEx and DistMult as the primary controlled pair, with targeted RotatE/TransE spot-checks, we evaluate seven benchmarks. On five standard KGs, ComplEx-vs-DistMult differences are modest but consistent under our recipe (+0.005 to +0.012 MRR), whereas CompGCN-style encoder effects vary more by dataset. On small KGs, decoder effects become the main diagnostic: Kinship shows a stable ComplEx advantage of +0.143 MRR (6 seeds), while UMLS favours ComplEx by +0.022 MRR in a clean 6-seed server rerun but reverses in an earlier provenance variant. We therefore treat small-KG decoder choice as recipe- and provenance-sensitive rather than as a fixed dataset winner. We further show that decoder choice interacts with encoder depth on WN18RR, and that under our recipe L=0 ComplEx on YAGO3-10 reaches 0.6971 +/- 0.0048 MRR at d=128. The result is a compact audit protocol: report matched decoder rows, log small-KG provenance, and sweep decoder x depth before making encoder-level claims.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 26 references that are in the map. First posted on arXiv 12 Jun 2026.

Builds on

21 of 26 references.

References · 21

  1. 2017Modeling Relational Data with Graph Convolutional Networkscited by 1,694
  2. 2014Embedding Entities and Relations for Learning and Inference in Knowledge Basescited by 932
  3. 2016Complex Embeddings for Simple Link Predictioncited by 902
  4. 2019RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Spacecited by 898
  5. 2017Convolutional 2D Knowledge Graph Embeddingscited by 730
  6. 2019Composition-based Multi-Relational Graph Convolutional Networkscited by 396
  7. 2019TuckER: Tensor Factorization for Knowledge Graph Completioncited by 290
  8. 2021Neural Bellman-Ford Networks: A General Graph Neural Network Framework for Link Predictioncited by 275
  9. 2018SimplE Embedding for Link Prediction in Knowledge Graphscited by 195
  10. 2018Canonical Tensor Decomposition for Knowledge Base Completioncited by 150
  11. 2019Learning Hierarchy-Aware Knowledge Graph Embeddings for Link Predictioncited by 133
  12. 2020CoDEx: A Comprehensive Knowledge Graph Completion Benchmarkcited by 102
9 more
  1. 2021Knowledge Graph Reasoning with Relational Digraphcited by 92
  2. 2017Knowledge Graph Completion via Complex Tensor Factorizationcited by 80
  3. 2019A Re-evaluation of Knowledge Graph Completion Methodscited by 73
  4. 2020Bringing Light Into the Dark: A Large-scale Evaluation of Knowledge Graph Embedding Models Under a Unified Frameworkcited by 70
  5. 2020HittER: Hierarchical Transformers for Knowledge Graph Embeddingscited by 54
  6. 2022A*Net: A Scalable Path-based Reasoning Approach for Knowledge Graphscited by 49
  7. 2020Realistic Re-evaluation of Knowledge Graph Completion Methods: An Experimental Studycited by 37
  8. 2022Rethinking Graph Convolutional Networks in Knowledge Graph Completioncited by 20
  9. 2018Data-dependent Learning of Symmetric/Antisymmetric Relations for Knowledge Base Completioncited by 3

Built on by

Not cited yet.