Conversational Question Answering on QReCC (test)
120EM (%)GroGU
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| GroGUType=BM25, LLM=QWEN2026.01 | 120 | 32.3 | 26.8 | 53.9 | 60.5 | |
| QWEN - Base SFTType=BM252026.01 | 110 | 30.8 | 25.7 | 51.6 | 56 | |
| QWEN - Base SFTType=ANCE2026.01 | 110 | 30.1 | 24.7 | 51.6 | 55.7 | |
| GroGUType=ANCE, LLM=QWEN2026.01 | 110 | 31.1 | 25.6 | 52.9 | 58.3 | |
| QWEN - OriginalType=BM252026.01 | 100 | 28.8 | 23.8 | 49.3 | 51.4 | |
| QWEN - OriginalType=ANCE2026.01 | 90 | 27.9 | 22.9 | 48.7 | 50.4 | |
| PHI - OriginalType=BM252026.01 | 40 | 32.9 | 25.7 | 54.2 | 59.2 | |
| GroGUType=BM25, LLM=PHI2026.01 | 40 | 34.1 | 26.5 | 56.5 | 62.6 | |
| PHI - OriginalType=ANCE2026.01 | 40 | 32 | 24.8 | 53.3 | 57.6 | |
| GroGUType=ANCE, LLM=PHI2026.01 | 40 | 33 | 25.6 | 55 | 60.1 | |
| PHI - Base SFTType=ANCE2026.01 | 30 | 31.5 | 24.4 | 52.6 | 57.5 | |
| PHI - Base SFTType=BM252026.01 | 20 | 31.9 | 24.8 | 53 | 58.1 |