Automated Scoring on NGSS-aligned modeling Item 3 (test)
82.7AccuracyCA-Selective
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CA-SelectiveBackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=Selective2026.06 | 82.7 | 0.826 | 80.5 | 81.5 | 80.9 | |
| CA-UniformBackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=Uniform2026.06 | 80.8 | 0.747 | 77.4 | 78.2 | 77.8 | |
| ViT + LoRABackbone=vit_base_patch16_224, Fine-tuning strategy=LoRA, Confidence-aware scoring mechanism=None2026.06 | 76.9 | 0.708 | 72.8 | 73.6 | 73.1 | |
| ViT (Frozen)Backbone=vit_base_patch16_224, Fine-tuning strategy=Frozen, Confidence-aware scoring mechanism=None2026.06 | 21.2 | -0.11 | 24.6 | 25.9 | 19.6 |