Identifying supporting dialogue evidence on Chen et al. (test)
77.8Hit@1Hierarchical
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Hierarchical2026.04 | 77.8 | 85.6 | 94.5 | 80.3 | |
| Drop+Hold2026.04 | 73.6 | 82.5 | 91.1 | 76.6 | |
| Leave-one-out2026.04 | 64.1 | 72.4 | 85 | 71.9 | |
| GradNorm2026.04 | 57.8 | 70.1 | 80.4 | 68.6 | |
| Similarity2026.04 | 51.1 | 67 | 73.8 | 63.9 |