Skill Retrieval on TheoremQA
77.4nDCG@1Qwen3-32B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-32BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 77.4 | 83.5 | |
| Llama-3.3-70BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 76 | 82.8 | |
| Qwen3-235BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 75.4 | 83.1 | |
| Mistral3.1-24BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 74.2 | 81.9 | |
| Qwen3-4BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 69.5 | 79 | |
| BGERetrieval Stage=First-stage2026.04 | 66.8 | 75.9 | |
| Llama-3.1-8BRetrieval Stage=Reranker, Candidate selection=top-50 candidates retrieved by BM252026.04 | 58.8 | 71.5 | |
| BM25Retrieval Stage=First-stage2026.04 | 57.2 | 69.2 | |
| HybridRetrieval Stage=First-stage2026.04 | 57.2 | 75.3 | |
| ContrieverRetrieval Stage=First-stage2026.04 | 52.1 | 63.8 | |
| TF-IDFRetrieval Stage=First-stage2026.04 | 41.4 | 54.4 |