Question Answering on StackExchange Q&A (test)
82.2Accuracy (Bio.)GTE-Base (w/ ERM)
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| GTE-Base (w/ ERM)Retriever=GTE-Base, ERM-augmented=true2026.02 | 82.2 | 80.9 | 75.1 | 79.9 | 76.9 | 84.6 | 73.6 | 79 | |
| Cohere (w/ ERM)Retriever=Cohere, ERM-augmented=true2026.02 | 81.9 | 82.3 | 77.8 | 81.2 | 78.7 | 86.9 | 74.9 | 80.5 | |
| BGE-Large (w/ ERM)Retriever=BGE-Large, ERM-augmented=true2026.02 | 81.2 | 80.6 | 73.7 | 76.9 | 76.4 | 81.3 | 73.1 | 77.6 | |
| CohereRetriever=Cohere, ERM-augmented=false2026.02 | 81.1 | 82.9 | 76.6 | 78.7 | 79.9 | 82.1 | 73.7 | 79.3 | |
| GTE-BaseRetriever=GTE-Base, ERM-augmented=false2026.02 | 80.5 | 81.6 | 74.4 | 77.5 | 75.3 | 79.7 | 73.1 | 77.4 | |
| BGE-LargeRetriever=BGE-Large, ERM-augmented=false2026.02 | 80.2 | 79.3 | 67.6 | 69.1 | 74.8 | 78.5 | 72.2 | 74.5 | |
| BM25 (w/ ERM)Retriever=BM25, ERM-augmented=true2026.02 | 79.9 | 77.8 | 73 | 76.1 | 76.4 | 80.5 | 72.5 | 76.6 | |
| BM25Retriever=BM25, ERM-augmented=false2026.02 | 71.8 | 74.5 | 69.8 | 74.8 | 71.4 | 76.4 | 69.5 | 72.6 |