Agentic Retrieval on BRIGHT-PRO Round 3 (test)
63.04Alpha-nDCGBGE-Reasoner-8B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| BGE-Reasoner-8BEvaluation Agent=GPT-5-mini2026.05 | 63.04 | 4.42 | 4.31 | |
| DIVER-4BEvaluation Agent=GPT-5-mini2026.05 | 53.08 | 4.38 | 4.29 | |
| GTE-7BEvaluation Agent=GPT-5-mini2026.05 | 52.68 | 4.33 | 4.23 | |
| Qwen3-8BEvaluation Agent=GPT-5-mini2026.05 | 51.75 | 4.26 | 4.1 | |
| DIVER-4B-1020Evaluation Agent=GPT-5-mini2026.05 | 51.56 | 4.33 | 4.16 | |
| BM25Evaluation Agent=GPT-5-mini2026.05 | 51.48 | 4.25 | 4.12 | |
| RTriever-4BEvaluation Agent=GPT-5-mini2026.05 | 50.79 | 4.37 | 4.25 | |
| GritLM (7B)Evaluation Agent=GPT-5-mini2026.05 | 47.05 | 4.24 | 4.07 | |
| ReasonIR-8BEvaluation Agent=GPT-5-mini2026.05 | 44.9 | 4.16 | 4.11 | |
| INSTRUCTOR-XLEvaluation Agent=GPT-5-mini2026.05 | 43.48 | 4.26 | 4.14 |