Science Question Answering on ARC-E (test)
79.5AccuracyFLAN-137B
Evaluation Results
| Method | Links | |
|---|---|---|
| FLAN-137Bsetting=zero-shot2026.03 | 79.5 | |
| Imagine-DeBERTa-v3-LKB=Synthetic VQA+, setting=zero-shot2026.03 | 79.1 | |
| Imagine-DeBERTa-v3-L (Retrieval)KB=Synthetic VQA+, inference_mode=Retrieval, setting=zero-shot2026.03 | 78.9 | |
| Imagine-DeBERTa-v3-LKB=Synthetic VQA, setting=zero-shot2026.03 | 76 | |
| CAR-DeBERTa-v3-LKB=AbsAT, setting=zero-shot2026.03 | 75.3 | |
| Z-LaVI (OPT-30B)setting=zero-shot2026.03 | 59.5 | |
| OPT-30Bsetting=zero-shot2026.03 | 58.2 | |
| Imagine-RoBERTa-LKB=Synthetic VQA, setting=zero-shot2026.03 | 57.9 | |
| CAR-RoBERTa-LKB=AbsAT, setting=zero-shot2026.03 | 57 | |
| Z-LaVI (BART-L)setting=zero-shot2026.03 | 56.1 | |
| Imagine-GPT-2-LKB=Synthetic VQA, setting=zero-shot2026.03 | 55.1 | |
| Z-LaVI (RoBERTa-L)setting=zero-shot2026.03 | 51.8 | |
| GPT-Neo-2.7Bsetting=zero-shot2026.03 | 49.6 | |
| GPT-J-6Bsetting=zero-shot2026.03 | 44.1 | |
| SMLMKB=*, setting=zero-shot2026.03 | 33.4 |