paper-with-me

Link Property Prediction 벤치마크

Link Property Prediction on ogbl-wikikg2

30개 결과 · ⬇ CSV · JSON

Test MRR

0.253 0.3745 0.4961 0.6177 0.7392 2014-12 2026-09 DistMult (500dim) — 0.3729 (2014-12-20) DistMult (100dim) — 0.3447 (2014-12-20) ComplEx (250dim) — 0.4027 (2016-06-20) ComplEx (50dim) — 0.3804 (2016-06-20) RotatE (250dim) — 0.4332 (2019-02-26) RotatE (50dim) — 0.253 (2019-02-26) AutoSF — 0.5458 (2019-04-26) PairRE (200dim) — 0.5208 (2020-11-07) NodePiece + AutoSF — 0.5703 (2021-06-23) ComplEx-N3-RP (100dim) — 0.6481 (2021-10-06) ComplEx-N3-RP (50dim) — 0.6364 (2021-10-06) Rot-Pro — 0.5602 (2021-10-27) RotPro — 0.4277 (2021-10-27) InterHT+ — 0.7293 (2022-02-10) InterHT+ (256dim) — 0.7257 (2022-02-10) InterHT — 0.6779 (2022-02-10) TranS — 0.6882 (2022-04-18) StarGraph + TripleRE — 0.7201 (2022-05-27) A*Net — 0.6711 (2022-06-07) AutoWeird — 0.7353 (2022-07-24) CompoundE3D — 0.7006 (2023-04-01) TransERR — 0.6359 (2023-06-26) RelEns — 0.7392 (2023-10-13) DistMult (500dim) — 0.3729 (2014-12-20) ComplEx (250dim) — 0.4027 (2016-06-20) RotatE (250dim) — 0.4332 (2019-02-26) AutoSF — 0.5458 (2019-04-26) NodePiece + AutoSF — 0.5703 (2021-06-23) ComplEx-N3-RP (100dim) — 0.6481 (2021-10-06) InterHT+ — 0.7293 (2022-02-10) AutoWeird — 0.7353 (2022-07-24) RelEns — 0.7392 (2023-10-13)
RankModel Test MRRExt. dataValidation MRRNumber of params PaperCodeYear
1 RelEns 0.7392 ± 0.0011No0.7509 ± 0.00092176767622 Relation-aware Ensemble Learning for Knowledge Graph Embedding lars-research/relens · LARS-research/RelEns 2023
2 AutoWeird 0.7353 ± 0.0006No0.7362 ± 0.000619215402 AutoWeird: Weird Translational Scoring Function Identified by Random Search 2022
3 StarGraph + TripleRE + Text 0.7305 ± 0.0010Yes0.7442 ± 0.00061927395330
4 InterHT+ 0.7293 ± 0.0018No0.7391 ± 0.0023156332770 InterHT: Knowledge Graph Embeddings by Interaction between Head and Tail Entities 2022
5 InterHT+ (256dim) 0.7257 ± 0.0018No0.7370 ± 0.0022148000738 InterHT: Knowledge Graph Embeddings by Interaction between Head and Tail Entities 2022
6 StarGraph + TripleRE 0.7201 ± 0.0011No0.7288 ± 0.000886762146 StarGraph: Knowledge Representation Learning based on Incomplete Two-hop Subgraph hzli-ucas/stargraph · hzli-ucas/StarGraph 2022
7 CompoundE3D 0.7006 ± 0.0011No0.7175 ± 0.0015750662700 Knowledge Graph Embedding with 3D Compound Geometric Transformations hughxiouge/CompoundE3D 2023
8 TranS 0.6882 ± 0.0019No0.6988 ± 0.000619215402 TranS: Transition-based Knowledge Graph Embedding with Synthetic Relation Representation 2022
9 InterHT 0.6779 ± 0.0018No0.6893 ± 0.001519215402 InterHT: Knowledge Graph Embeddings by Interaction between Head and Tail Entities 2022
10 A*Net 0.6711 ± 0.0045No0.6796 ± 0.00706831201
11 A*Net 0.6711 ± 0.0045No0.6796 ± 0.00706831201 A*Net: A Scalable Path-based Reasoning Approach for Knowledge Graphs deepgraphlearning/astarnet · DeepGraphLearning/AStarNet 2022
12 TripleRE + NodePiece 0.6582 ± 0.0020No0.6616 ± 0.00187289002
13 ComplEx-N3-RP (100dim) 0.64810.6701500334800 Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations facebookresearch/ssl-relation-prediction 2021
14 ComplEx-RP (50dim) 0.6392 ± 0.0045No0.6561 ± 0.0070250167400
15 ComplEx-N3-RP (50dim) 0.63640.6594250167400 Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations facebookresearch/ssl-relation-prediction 2021
16 TransERR 0.6359 ± 0.0020No0.6518 ± 0.0012500441802 TransERR: Translation-based Knowledge Graph Embedding via Efficient Relation Rotation dellixx/transerr · dellixx/TransERR 2023
17 tripleRE 0.5794 ± 0.0020No0.6045 ± 0.0024500763337
18 NodePiece + AutoSF 0.5703 ± 0.0035No0.5806 ± 0.00476860602 NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs migalkin/NodePiece · migalkin/NodePiece · pykeen/ilpc2022 · +1 2021
19 Rot-Pro 0.5602 ± 0.0016No0.5740 ± 0.00081000669602 Rot-Pro: Modeling Transitivity by Projection in Knowledge Graph Embedding tewiSong/Rot-Pro · Tigter/ogblwikikg2-RotPro 2021
20 AutoSF 0.5458 ± 0.0052No0.5510 ± 0.0063500227800 AutoSF: Searching Scoring Functions for Knowledge Graph Embedding AutoML-4Paradigm/ERAS · AutoML-4Paradigm/S2S · AutoML-Research/S2S 2019
1–20 / 30 다음 → 페이지당 10 20 50 100