Ranking Uncertainty Quantification on MovieLens 1M
0.1576Kendall's Tau@5EviRank
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| EviRankBackbone=Qwen2.52026.06 | 0.1576 | 0.6002 | 0.1489 | 0.5873 | |
| EviRankBackbone=Mistral2026.06 | 0.151 | 0.5932 | 0.1373 | 0.5748 | |
| Verb. 1S top-1Backbone=Qwen2.52026.06 | 0.1482 | 0.5855 | 0.1257 | 0.5597 | |
| EviRankBackbone=Llama32026.06 | 0.1478 | 0.5805 | 0.1369 | 0.5671 | |
| Label Prob.Backbone=Qwen2.52026.06 | 0.1426 | 0.5793 | 0.1205 | 0.5593 | |
| Semantic Unc.Backbone=Qwen2.52026.06 | 0.141 | 0.5835 | 0.1172 | 0.5572 | |
| Verb. 1S top-1Backbone=Llama32026.06 | 0.1369 | 0.5598 | 0.126 | 0.5471 | |
| Label Prob.Backbone=Llama32026.06 | 0.1283 | 0.5565 | 0.1143 | 0.5343 | |
| Semantic Unc.Backbone=Llama32026.06 | 0.1162 | 0.5587 | 0.1072 | 0.538 | |
| Verb. 1S top-1Backbone=Mistral2026.06 | 0.1123 | 0.5617 | 0.093 | 0.5428 | |
| Semantic Unc.Backbone=Mistral2026.06 | 0.1089 | 0.5687 | 0.0898 | 0.5469 | |
| Label Prob.Backbone=Mistral2026.06 | 0.101 | 0.5631 | 0.0826 | 0.5441 |