Semantic Similarity On Sick
Metrics
MSE
Pearson Correlation
Spearman Correlation
Results
Performance results of various models on this benchmark
Model Name | MSE | Pearson Correlation | Spearman Correlation | Paper Title | Repository |
|---|---|---|---|---|---|
| Bidirectional LSTM (Tai et al., 2015) | 0.2736 | 0.8567 | 0.7966 | Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks | |
| LSTM (Tai et al., 2015) | 0.2831 | 0.8528 | 0.7911 | Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks | |
| combine-skip (Kiros et al., 2015) | 0.2687 | 0.8584 | 0.7916 | Skip-Thought Vectors | |
| Dependency Tree-LSTM (Tai et al., 2015) | 0.2532 | 0.8676 | 0.8083 | Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks | |
| Doc2VecC | 0.3053 | 0.8381 | 0.7621 | Efficient Vector Representation for Documents through Corruption |
0 of 5 row(s) selected.