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

Malicious Python package detection on PyPI-based dataset Large Import Amount, N=487 (test)

0.9584Accuracy

H2GLM

Updated 18d ago

Evaluation Results

Method	Links
H2GLM 2026.07		0.9584	0.9514	0.9662	0.9588
HERCULE 2026.07		0.8511	0.785	0.9671	0.8666
MPHunter 2026.07		0.7972	0.748	0.772	0.7598
Warehouse 2026.07		0.7444	0.9161	0.538	0.6779
Qwen3-14b 2026.07		0.6783	0.9539	0.3747	0.538
OSSGadget 2026.07		0.5492	0.442	0.84	0.5792
DeepSeek-R1-70b 2026.07		0.4742	0.4638	0.3307	0.3861
Naive Bayes 2026.07		0.4538	0.4148	0.29	0.3414
Bandit4Mal 2026.07		0.3614	0.2816	0.165	0.2081
Qwen2.5-7b 2026.07		0.354	0.29	0.2016	0.2378
Clustering 2026.07		0.3465	0.1505	0.0661	0.0919
Llama3-8b 2026.07		0.208	0.1753	0.1576	0.166