Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Machine-generated code detection on SemEval Task 13 Subtask A 2026 official (test)

78.9Macro F1

SALSA Tuned (Qwen2.5-72B-Instruct)

28.56441.63254.767.768Apr 23, 2026May 3, 2026May 13, 2026May 23, 2026Jun 2, 2026Jun 12, 2026Jun 23, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
78.9
2026.06
76
2026.06
73
2026.04
69.753
2026.04
64.625
2026.04
64.435
2026.04
63.818
2026.04
59.628
2026.06
59.1
2026.04
57.799
2026.04
56.363
2026.04
54.776
2026.04
54.498
2026.04
54.183
2026.04
51.694
2026.06
50.8
50
2026.04
49.474
2026.04
48.006
2026.06
45
2026.06
44.9
2026.06
43.6
30.53
2026.06
30.5