Science Question Answering on QASC (test)
73.5AccuracyImagine-DeBERTa-v3-L
Evaluation results
| Method | Links | |
|---|---|---|
| Imagine-DeBERTa-v3-LKB=Synthetic VQA+, setting=zero-shot2026.03 | 73.5 | |
| Imagine-DeBERTa-v3-L (Retrieval)KB=Synthetic VQA+, inference_mode=Retrieval, setting=zero-shot2026.03 | 73.5 | |
| Imagine-DeBERTa-v3-LKB=Synthetic VQA, setting=zero-shot2026.03 | 72.4 | |
| CAR-DeBERTa-v3-LKB=AbsAT, setting=zero-shot2026.03 | 70 | |
| Imagine-RoBERTa-LKB=Synthetic VQA, setting=zero-shot2026.03 | 57.1 | |
| CAR-RoBERTa-LKB=AbsAT, setting=zero-shot2026.03 | 56.7 | |
| Imagine-GPT-2-LKB=Synthetic VQA, setting=zero-shot2026.03 | 46.5 | |
| Z-LaVI (OPT-30B)setting=zero-shot2026.03 | 42.1 | |
| OPT-30Bsetting=zero-shot2026.03 | 39.7 | |
| GPT-J-6Bsetting=zero-shot2026.03 | 36.3 | |
| GPT-Neo-2.7Bsetting=zero-shot2026.03 | 29.6 | |
| Z-LaVI (BART-L)setting=zero-shot2026.03 | 27.3 | |
| Z-LaVI (RoBERTa-L)setting=zero-shot2026.03 | 27.2 | |
| SMLMKB=*, setting=zero-shot2026.03 | 26.6 |