LLM-generated text detection on HC3 GPT-5.4 generated text benchmark leaderboard · SOTA2 Research

LLM-generated text detection on HC3 GPT-5.4 generated text

0.974ROC AUCDetectGPT-d

Evaluation Results

Method	Links
DetectGPT-d2026.07		0.974	0.976	0.928	0.907	0.926	0.911	0.945	0.938
JCASEMCD2026.07		0.969	0.97	0.915	0.973	0.92	0.969	0.867	0.941
JCELLMCD2026.07		0.963	0.957	0.913	0.905	0.912	0.906	0.919	0.925
DetectGPT-z2026.07		0.958	0.949	0.921	0.893	0.919	0.898	0.945	0.926
Likelihood2026.07		0.956	0.937	0.928	0.897	0.926	0.902	0.955	0.929
LogRank2026.07		0.954	0.934	0.932	0.894	0.93	0.901	0.966	0.93
Rank2026.07		0.881	0.843	0.84	0.792	0.833	0.808	0.874	0.839
RoBERTa-large2026.07		0.235	0.353	0.667	0	0.5	0.5	1	0.465
RoBERTa-base2026.07		0.231	0.351	0.667	0	0.5	0.5	1	0.464
Entropy2026.07		0.022	0.309	0.667	0	0.5	0.5	1	0.428