Question Answering on Aggregate of 4 Benchmarks (test)
27.34Delta EMVOI
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| VOIComparison baseline=Search-o1, Retrieval Backend=Search-R1, Queries=question-only, top_k=5, Audit=hard dual-budget2026.05 | 27.34 | 23.6 | 29.93 | 26.25 | |
| VOIComparison baseline=AFlow, Retrieval Backend=Search-R1, Queries=question-only, top_k=5, Audit=hard dual-budget2026.05 | 24.56 | 21 | 30.21 | 25.56 | |
| VOIComparison baseline=BATS, Retrieval Backend=Search-R1, Queries=question-only, top_k=5, Audit=hard dual-budget2026.05 | 10.81 | 4.37 | 12.19 | 4.39 | |
| VOIComparison baseline=BAVT, Retrieval Backend=Search-R1, Queries=question-only, top_k=5, Audit=hard dual-budget2026.05 | 4.92 | 3.82 | 5.3 | 4.23 |