Question Answering on QASC (Accuracy, Normalized Accuracy)
98.9AccuracyFine-tuned
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Fine-tunedBackbone=T5-large2026.06 | 98.9 | — | |
| ReTeXBackbone=T5-large, Merging Strategy=Dynamic model merging2026.06 | 97.4 | 98.5 | |
| TIES-MergingBackbone=T5-large, Merging Strategy=Static model merging2026.06 | 96.2 | 97.3 | |
| DaWinBackbone=T5-large, Merging Strategy=Dynamic model merging2026.06 | 96.1 | 97.2 | |
| Task ArithmeticBackbone=T5-large, Merging Strategy=Static model merging2026.06 | 96 | 97.1 | |
| Twin-MergingBackbone=T5-large, Merging Strategy=Dynamic model merging2026.06 | 96 | 97.1 | |
| WEMoEBackbone=T5-large, Merging Strategy=Dynamic model merging2026.06 | 95.4 | 96.5 | |
| MoW-MergingBackbone=T5-large, Merging Strategy=Dynamic model merging2026.06 | 90.2 | 91.2 | |
| Weight AveragingBackbone=T5-large, Merging Strategy=Static model merging2026.06 | 82.6 | 83.5 | |
| AdaMergingBackbone=T5-large, Merging Strategy=Static model merging2026.06 | 76 | 76.8 | |
| Consensus TABackbone=T5-large, Merging Strategy=Static model merging2026.06 | 67.4 | 68.1 |