Retrieval on Bird
96.1Recall@10Baseline
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| BaselineModel=stella2026.03 | 96.1 | — | — | — | — | |
| DCTRModel=bge-small, vote_k=2, n_groups=2, expand_groups=true2026.03 | 94.2 | — | — | — | — | |
| DCTRModel=stella, vote_k=2, n_groups=2, expand_groups=true2026.03 | 93.6 | — | — | — | — | |
| BaselineModel=bge-small2026.03 | 90.9 | — | — | — | — | |
| DCTRModel=e5-small, vote_k=2, n_groups=2, expand_groups=true2026.03 | 90.4 | — | — | — | — | |
| DCTRModel=stella, vote_k=2, n_groups=22026.03 | 89.9 | — | — | — | — | |
| DCTRModel=bge-small, vote_k=2, n_groups=22026.03 | 89.3 | — | — | — | — | |
| DCTRModel=e5-small, vote_k=2, n_groups=22026.03 | 88.2 | — | — | — | — | |
| BaselineModel=e5-small2026.03 | 85 | — | — | — | — | |
| ARMBackbone=Llama3.1-8B-Instruct, #calls=1, Avg #obj.=5.002025.01 | — | 42.7 | 96.5 | 56 | 92.7 | |
| ReActBackbone=Llama3.1-8B-It, #calls=5.26, Avg #obj.=17.32025.01 | — | 15 | 96.7 | 24.5 | 93.5 | |
| ReActBackbone=GPT4o-mini, #calls=3.16, Avg #obj.=11.42025.01 | — | 25.1 | 97 | 37.8 | 93.3 |