Results of the full model on narration test data. We show the mean and standard deviation of bootstrapped scores, pooled over four training runs (chance recall@10 = 10%; chance triplet accuracy = 50%).
| R@10 (fixed) . | R@10 (jitter) . | Triplet Acc . |
|---|---|---|
| 0.73 ± 0.05 | 0.73 ± 0.04 | 0.91 ± 0.01 |
| R@10 (fixed) . | R@10 (jitter) . | Triplet Acc . |
|---|---|---|
| 0.73 ± 0.05 | 0.73 ± 0.04 | 0.91 ± 0.01 |