Top-K Recommendation Evaluation Consistency on Coat
0.8741Pearson CorrelationLLM Judge
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| LLM JudgeTarget Metric=Recall@3, Backbone=Qwen3-30B-A3B2026.06 | 0.8741 | 0.9286 | 0.7857 | 49.25 | |
| LLM JudgeTarget Metric=Recall@5, Backbone=Qwen3-30B-A3B2026.06 | 0.857 | 0.5241 | 0.4001 | 51.64 | |
| LLM JudgeTarget Metric=NDCG@5, Backbone=Qwen3-30B-A3B2026.06 | 0.8561 | 0.7619 | 0.5714 | 133.93 | |
| LLM JudgeTarget Metric=NDCG@10, Backbone=Qwen3-30B-A3B2026.06 | 0.8272 | 0.8333 | 0.7143 | 112.85 | |
| LLM JudgeTarget Metric=Recall@10, Backbone=Qwen3-30B-A3B2026.06 | 0.762 | 0.381 | 0.2143 | 85.11 | |
| LLM JudgeTarget Metric=NDCG@3, Backbone=Qwen3-30B-A3B2026.06 | 0.7386 | 0.8095 | 0.6429 | 151.43 | |
| Normal TestingTarget Metric=Recall@32026.06 | 0.3815 | 0.2381 | 0.2143 | — | |
| Normal TestingTarget Metric=Recall@52026.06 | 0.3406 | 0.0952 | 0.0714 | — | |
| Normal TestingTarget Metric=Recall@102026.06 | -0.0891 | -0.1667 | -0.0714 | — | |
| Normal TestingTarget Metric=NDCG@102026.06 | -0.3012 | -0.4524 | -0.3571 | — | |
| Normal TestingTarget Metric=NDCG@52026.06 | -0.4831 | -0.4762 | -0.2857 | — | |
| Normal TestingTarget Metric=NDCG@32026.06 | -0.7757 | -0.7619 | -0.5714 | — |