Machine-Generated Text Detection on M4

87.28TP @ 20%Likelihood

Evaluation results

Method
Likelihood2025.05	87.28	84.17	78.68	83.52	70.04	80.1	58.12	72.89	48.28	64.88	36.04	52.83
Likelihood2025.05	87.28	84.17	78.68	83.52	70.04	80.1	58.12	72.89	48.28	64.88	36.04	52.83
Log-Rank2025.05	87.24	84.32	78.8	83.72	70.12	80.12	56.68	71.66	48.48	64.99	38.56	55.51
Log-Rank2025.05	87.24	84.32	78.8	83.72	70.12	80.12	56.68	71.66	48.48	64.99	38.56	55.51
Binoculars2025.05	83.24	82.4	80	84.41	75.72	83.87	71.2	82.27	66.68	79.59	62.72	76.88
Binoculars2025.05	83.2	81.87	79.44	83.87	74.8	83.19	69.72	81.21	64.52	77.96	55.36	71.03
Binoculars2025.05	83.2	81.87	79.44	83.87	74.8	83.19	69.72	81.21	64.52	77.96	55.36	71.03
Binoculars2025.05	83.2	81.87	79.44	83.87	74.8	83.19	69.72	81.21	64.52	77.96	55.36	71.03
Binoculars2025.05	83.08	82.03	78.84	83.96	75.72	83.87	71.2	82.27	66.68	79.59	62.72	76.88
Binoculars2025.05	83.08	82.03	78.84	83.96	75.72	83.87	71.2	82.27	66.68	79.59	62.72	76.88
Log-Rank2025.05	82.52	81.49	71.84	79.03	62.04	74.28	50.76	66.47	42.96	59.68	35.32	52.02
Log-Rank2025.05	82.52	81.49	71.84	79.03	62.04	74.28	50.76	66.47	42.96	59.68	35.32	52.02
Likelihood2025.05	81.48	80.87	72.76	79.62	65.12	76.54	54.72	69.85	43.92	60.61	35.96	52.7
Likelihood2025.05	81.48	80.87	72.76	79.62	65.12	76.54	54.72	69.85	43.92	60.61	35.96	52.7
Fast-DetectGPT2025.05	79.56	80.35	75.24	81.5	71.72	81.33	67.36	79.55	62.44	76.33	58.64	73.67
Fast-DetectGPT2025.05	79.56	80.35	75.24	81.5	71.72	81.33	67.36	79.55	62.44	76.33	58.64	73.67
Fast-DetectGPT2025.05	79.56	80.35	75.24	81.5	71.72	81.33	67.36	79.55	62.44	76.33	58.64	73.67
Fast-DetectGPT2025.05	78.56	79.13	74.44	80.72	70.04	80.03	65.32	78.08	60.56	74.97	54.84	70.6
Fast-DetectGPT2025.05	78.56	79.13	74.44	80.72	70.04	80.03	65.32	78.08	60.56	74.97	54.84	70.6
Fast-DetectGPT2025.05	78.56	79.13	74.44	80.72	70.04	80.03	65.32	78.08	60.56	74.97	54.84	70.6
OpenAI-D2025.05	67.24	71.43	59.28	69.84	52.08	66.45	45.04	61.28	38.92	55.73	35.04	51.76
OpenAI-D2025.05	67.24	71.43	59.28	69.84	52.08	66.45	45.04	61.28	38.92	55.73	35.04	51.76
Entropy2025.05	64.8	70.59	45.48	58.71	28.44	42.82	14.84	25.53	9.88	17.87	6.48	12.14
Entropy2025.05	64.8	70.59	45.48	58.71	28.44	42.82	14.84	25.53	9.88	17.87	6.48	12.14
OpenAI-D2025.05	64	69.53	54.12	65.95	49.6	64.17	42.6	58.92	38.56	55.26	35.88	52.63
OpenAI-D2025.05	64	69.53	54.12	65.95	49.6	64.17	42.6	58.92	38.56	55.26	35.88	52.63
DetectGPT2025.05	62	68.19	46.28	59.35	31.52	46.12	18.36	30.53	11.52	20.49	7.08	13.17
DetectGPT2025.05	62	68.19	46.28	59.35	31.52	46.12	18.36	30.53	11.52	20.49	7.08	13.17
DetectGPT2025.05	61.2	67.53	44.68	57.77	31.12	45.72	16.96	28.51	10.08	18.15	7.28	13.51
DetectGPT2025.05	61.2	67.53	44.68	57.77	31.12	45.72	16.96	28.51	10.08	18.15	7.28	13.51
Entropy2025.05	57.12	64.5	35.28	48.53	25.12	38.61	14	24.14	10.48	18.8	8.48	15.57
Entropy2025.05	57.12	64.5	35.28	48.53	25.12	38.61	14	24.14	10.48	18.8	8.48	15.57