Trait-based Automatic Essay Scoring on ASAP++ Prompt 1 (test)
0.59QWK (Content)BigBird-CORAL
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| BigBird-CORALEvaluation Protocol=Fine-tuning2026.02 | 0.59 | 0.53 | 0.61 | 0.47 | 0.48 | 0.71 | 0.68 | 0.76 | 0.65 | 0.67 | |
| BigBird-RMSEEvaluation Protocol=Fine-tuning2026.02 | 0.53 | 0.49 | 0.56 | 0.47 | 0.45 | 0.73 | 0.74 | 0.77 | 0.71 | 0.71 | |
| Ministral-3 8BEvaluation Protocol=Sampling for LLMs2026.02 | 0.49 | 0.43 | 0.32 | 0.19 | 0.33 | 0.69 | 0.63 | 0.7 | 0.67 | 0.7 | |
| BigBird-nominalEvaluation Protocol=Fine-tuning2026.02 | 0.39 | 0.44 | 0.43 | 0.33 | 0.33 | 0.73 | 0.74 | 0.77 | 0.71 | 0.71 | |
| Llama-3.1 8BEvaluation Protocol=Sampling for LLMs2026.02 | 0.22 | 0.25 | 0.26 | 0.2 | 0.22 | 0.49 | 0.53 | 0.61 | 0.52 | 0.6 | |
| Gemma-3 12BEvaluation Protocol=Sampling for LLMs2026.02 | 0.09 | 0.16 | 0.13 | 0.13 | 0.19 | 0.59 | 0.65 | 0.64 | 0.62 | 0.63 |