Hyper-Relational Link Prediction on JFFI V1
0.3683MRR (H/T)THOR
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| THORSetting=Inductive2026.02 | 0.3683 | — | 0.2638 | — | |
| THORInductive setting=Fully-inductive2026.02 | 0.3683 | — | — | 0.2638 | |
| TRIXInductive setting=Fully-inductive2026.02 | 0.3122 | — | — | — | |
| ULTRAInductive setting=Fully-inductive2026.02 | 0.3073 | — | — | — | |
| MOTIFInductive setting=Fully-inductive2026.02 | 0.2765 | — | — | — | |
| MetaNIRInductive setting=Fully-inductive2026.02 | 0.249 | — | — | 0.1914 | |
| INGRAMInductive setting=Fully-inductive2026.02 | 0.164 | — | — | — | |
| MAYPLInductive setting=Fully-inductive2026.02 | 0.1331 | — | — | 0.0967 | |
| KG-ICLInductive setting=Fully-inductive2026.02 | 0.0746 | — | — | — | |
| HAHESetting=Inductive2026.02 | 0.0677 | — | 0.0549 | — | |
| GRANSetting=Inductive2026.02 | 0.0487 | — | 0.0373 | — | |
| StarESetting=Inductive2026.02 | 0.0327 | — | — | — | |
| NS-HARTInductive setting=Semi-inductive2026.02 | 0.0211 | — | — | 0.0223 | |
| Neural LPInductive setting=Rule-based2026.02 | 0.017 | — | — | — | |
| DRUMInductive setting=Rule-based2026.02 | 0.0093 | — | — | — | |
| RMPIInductive setting=Fully-inductive2026.02 | 0.0047 | — | — | — | |
| NBFNetInductive setting=Semi-inductive2026.02 | 0.0029 | — | — | — | |
| QBLP (w/o pretrain)Inductive setting=Semi-inductive, Pre-training status=Without pre-training2026.02 | 0.0023 | — | — | — | |
| GRANEvaluation Protocol=transductive2026.02 | — | 0.3027 | 0.2662 | — | |
| HAHEEvaluation Protocol=transductive2026.02 | — | 0.3256 | 0.2598 | — | |
| StarEEvaluation Protocol=transductive2026.02 | — | 0.3345 | — | — | |
| THOREvaluation Protocol=transductive2026.02 | — | 0.3683 | 0.2638 | — |