Word Similarity on WS-353 SIM
0.66Spearman CorrelationW2V*text8
Evaluation results
| Method | Links | ||
|---|---|---|---|
| W2V*text8Embedding Method=W2V, Training Corpus=text82018.10 | 0.66 | — | |
| FastTextHardware=CPU2026.05 | 0.601 | 0.42 | |
| GloVe+GTEmbedding Method=GloVe, Transform=Gaussian Transform, Refinement Corpus=text82018.10 | 0.6 | — | |
| EllEmbedding Method=Ell, Training Corpus=ukWaC and WaCkypedia2018.10 | 0.6 | — | |
| Word2VecHardware=CPU2026.05 | 0.591 | 0.415 | |
| GloVeEmbedding Method=GloVe2018.10 | 0.57 | — | |
| GloVeHardware=CPU2026.05 | 0.514 | 0.355 | |
| GloVe*text8Embedding Method=GloVe, Training Corpus=text82018.10 | 0.51 | — | |
| W2GEmbedding Method=W2G, Training Corpus=ukWaC and WaCkypedia2018.10 | 0.48 | — | |
| Omni TM-AEHardware=CPU2026.05 | 0.478 | 0.344 | |
| FastOmniTMAEHardware=CPU2026.05 | 0.47 | 0.343 | |
| FastOmniTMAEHardware=CUDA2026.05 | 0.469 | 0.347 |