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

Code Hallucination Detection on BigCode

0.77OP

o3-mini

0.13560.30030.4650.6297Apr 9, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.04
0.770.480.5
2026.04
0.640.120.15
2026.04
0.620.30.36
2026.04
0.570.110.14
2026.04
0.560.160.14
2026.04
0.470.090.1
2026.04
0.450.090.06
2026.04
0.340.20.24
2026.04
0.320.20.24
2026.04
0.280.160.14
2026.04
0.240.090.08
2026.04
0.240.120.11
2026.04
0.230.060.06
2026.04
0.220.070.03
2026.04
0.190.040.06
2026.04
0.160.050.03