Machine-generated code detection on SemEval-2026 Task 13 Subtask A (train)
99.6Macro F1SALSA Tuned (Qwen2.5-72B-Instruct)
Evaluation results
| Method | Links | |
|---|---|---|
| SALSA Tuned (Qwen2.5-72B-Instruct)Tuning strategy=Fine-tuned, Model backbone=Qwen2.5-72B-Instruct2026.06 | 99.6 | |
| SALSA Tuned (Qwen3-32B)Tuning strategy=Fine-tuned, Model backbone=Qwen3-32B2026.06 | 99.4 | |
| SALSA Tuned (Qwen3-8B)Tuning strategy=Fine-tuned, Model backbone=Qwen3-8B2026.06 | 99.1 | |
| SALSA Tuned (Qwen3-32B official)Tuning strategy=Fine-tuned, Model backbone=Qwen3-32B, Submission status=Official Kaggle submission2026.06 | 98.2 | |
| SALSA Zero-shot (Qwen3-Next-80B-A3B-Instruct)Tuning strategy=Zero-shot - without tuning, Model backbone=Qwen3-Next-80B-A3B-Instruct2026.06 | 66.2 | |
| SALSA Zero-shot (Qwen3-32B)Tuning strategy=Zero-shot - without tuning, Model backbone=Qwen3-32B2026.06 | 52.8 | |
| SALSA Zero-shot (Qwen3-8B)Tuning strategy=Zero-shot - without tuning, Model backbone=Qwen3-8B2026.06 | 52.3 | |
| SALSA Zero-shot (Qwen2.5-72B-Instruct)Tuning strategy=Zero-shot - without tuning, Model backbone=Qwen2.5-72B-Instruct2026.06 | 35.8 |