Table 5. Summary statistics of detection scores from Scenario 2: AI-assisted abstracts.

Statistic	GPTZero (%)	ZeroGPT (%)	DetectGPT (%)
Mean (SD)	37.65 (39.99)	20.92 (29.24)	52.36 (47.40)
• By author categories
• Native	30.68 (35.29)	21.91 (27.79)	54.40 (46.79)
• Non-native	44.61 (43.33)	19.94 (30.79)	50.32 (48.24)
• By disciplines
• Tech & Engineering	41.04 (40.21)	10.99 (21.30)	53.90 (48.41)
• Social Sciences	35.21 (38.12)	20.06 (33.15)	52.10 (46.92)
• Interdisciplinary	36.69 (42.16)	31.73 (28.77)	51.08 (47.81)
• By LLMs
• ChatGPT o1	19.79 (30.51)	10.04 (21.50)	29.47 (21.50)
• Gemini 2.0 Pro	55.50 (40.55)	31.80 (31.93)	75.25 (31.93)
Q1, Median, Q3	2.00, 12.50, 80.25	0.00, 0.00, 40.51	0.00, 80.50, 100.00
• By author categories
• Native	3.00, 9.50, 58.00	0.00, 0.00, 44.84	0.00, 81.50, 100.00
• Non-native	2.00, 22.50, 99.25	0.00, 0.00, 36.38	0.00, 80.50, 100.00
• By disciplines
• Tech & Engineering	2.00, 26.50, 80.25	0.00, 0.00, 5.85	0.00, 80.50, 100.00
• Social Sciences	2.00, 14.00, 74.25	0.00, 0.00, 34.72	0.00, 79.50, 100.00
• Interdisciplinary	3.00, 9.00, 90.25	0.00, 30.48, 52.96	0.00, 83.50, 100.00
• By LLMs
• ChatGPT o1	1.00, 5.00, 25.75	0.00, 0.00, 0.00	0.00, 0.00, 0.00
• Gemini 2.0 Pro	9.00, 58.00, 100.00	0.00, 29.34, 53.06	0.00, 78.50, 91.00
Min, Max	0.00, 100.00	0.00, 100.00	0.00, 100.00
• By author categories
• Native	0.00, 100.00	0.00, 100.00	0.00, 100.00
• Non-native	0.00, 100.00	0.00, 100.00	0.00, 100.00
• By disciplines
• Tech & Engineering	0.00, 100.00	0.00, 79.43	0.00, 100.00
• Social Sciences	0.00, 100.00	0.00, 100.00	0.00, 100.00
• Interdisciplinary	0.00, 100.00	0.00, 96.64	0.00, 100.00
• By LLMs
• ChatGPT o1	0.00, 100.00	0.00, 100.00	0.00, 100.00
• Gemini 2.0 Pro	0.00, 100.00	0.00, 100.00	0.00, 100.00