notesum.ai

Published at November 14

How Good is ChatGPT at Audiovisual Deepfake Detection: A Comparative Study of ChatGPT, AI Models and Human Perception

cs.CV

cs.AI

cs.HC

cs.LG

cs.MM

Released Date: November 14, 2024

Authors: Sahibzada Adil Shahzad¹, Ammarah Hashmi¹, Yan-Tsung Peng¹, Yu Tsao¹, Hsin-Min Wang¹

Aff.: ¹Institution not provided

Arxiv: http://arxiv.org/abs/2411.09266v1

Prompt	Precision	Recall	F1 Score	Rejection Rate	Accuracy
P1	0.583	0.368	0.451	7.50	54.0
P2	0.625	0.250	0.357	2.50	53.8
P3	0.625	0.250	0.357	0.00	55.0
P4	0.571	0.600	0.585	0.00	57.5
P5	0.571	0.200	0.300	0.00	52.5
P6	0.607	0.850	0.706	0.00	65.0
P7	0.600	0.900	0.720	0.00	65.0