Science Question Answering on GPQA diamond en
32.32AccuracyBaseline
Evaluation Results
| Method | Links | |
|---|---|---|
| BaselineN=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.04 | 32.32 | |
| TLPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.04 | 31.66 | |
| ORPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.04 | 31.19 | |
| DPON=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.04 | 31.09 | |
| SFTN=16, setting=English occurrence as language confusion, aggregation=Averaged across four models and four target languages2026.04 | 27.87 |