Student Writing Feedback Generation on SEFORA Single-reference subset
38.7F1 ScoreQwen_72B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen_72Bshot-count=zero-shot, rubric=included, guidance=guided, prompt template=constrained2026.06 | 38.7 | |
| GPT-5.1shot-count=zero-shot, rubric=included, guidance=guided, prompt template=constrained2026.06 | 37.1 | |
| Mistral_7Bshot-count=zero-shot, rubric=included, guidance=guided, prompt template=constrained2026.06 | 31.1 | |
| Llama_8Bshot-count=zero-shot, rubric=included, guidance=guided, prompt template=constrained2026.06 | 29 |