AI-Generated Text Detection on Cross-genre (test) benchmark leaderboard · SOTA2 Research

AI-Generated Text Detection on Cross-genre (test)

97.5OALog-Likelihood

Evaluation Results

Method	Links
Log-Likelihood2024.06		97.5	96.5	1.03	6,077.26
Log-Rank2024.06		97.5	95.5	2.05	6,054.48
GLTR2024.06		97.5	67.5	30.77	5,476.04
BERT2024.06		95.5	75	21.47	5,619.45
RDrop2024.06		95	10	89.47	3,155.59
SCRN2024.06		94.5	71	24.87	4,419.16
ChatGPT-Detector2024.06		93	73	21.51	5,827.88
RMLM2024.06		92	72.5	21.2	5,440.61
DeBERTa2024.06		90	0	100	2,763.66
SeqXGPT2024.06		88	65.5	25.57	5,776.35
Flooding2024.06		87.5	0	100	2,733.18
RanMASK2024.06		87	75	13.79	5,433.82
SeqXGPT2024.06		85.5	0	100	2,712.97
RoBERTa2024.06		83	59	28.92	5,522.05
Flooding2024.06		82.5	58	29.7	5,447.84
RoBERTa2024.06		82	0	100	2,655.43
Entropy2024.06		77.5	0	100	2,783.14
DeBERTa2024.06		77.5	53.5	30.97	5,329.64
Entropy2024.06		74	34	54.05	5,352.47
RDrop2024.06		73	65	10.96	5,973.84
SCRN2024.06		70.5	54.5	22.7	5,725.79
RanMASK2024.06		67	2	97.01	2,667.19
Log-Rank2024.06		64.5	0	100	2,734.98
Log-Likelihood2024.06		62	0	100	2,700.46
ChatGPT-Detector2024.06		58.5	0	100	2,606.75
RMLM2024.06		58.5	9.5	83.76	3,397.99
BERT2024.06		57	0	100	2,692.61
GLTR2024.06		50.5	0	100	2,696.8
RoBERTa2024.06		0.825	0.31	0.6242	17,991.85
SCRN2024.06		0.8225	0.555	0.3253	23,179.41
RanMASK2024.06		0.77	0.3025	0.6071	15,757.8
RMLM2024.06		0.7525	0.4008	0.4673	20,728.78