Automated Scoring on NGSS-aligned modeling Item 1 (test)
79.2AccuracyViT + LoRA
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| ViT + LoRABackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=None2026.06 | 79.2 | 0.796 | 77.4 | 75.8 | 76.5 | |
| CA-UniformBackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=Uniform2026.06 | 79.2 | 0.796 | 77.4 | 75.8 | 76.5 | |
| CA-SelectiveBackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=Selective2026.06 | 79.2 | 0.796 | 77.4 | 75.8 | 76.5 | |
| ViT (Frozen)Backbone=vit_base_patch16_224, Fine-tuning strategy=Frozen, Confidence-aware scoring mechanism=None2026.06 | 43.8 | -0.048 | 31.4 | 35 | 25.7 |