Multimodal Deception Detection on MM-DeceptionBench benchmark leaderboard · SOTA2 Research

Multimodal Deception Detection on MM-DeceptionBench

0.822AccuracyDebate with images

Evaluation Results

Method	Links
Debate with images2025.11		0.822	0.52	0.88
CoT prompt2025.11		0.805	0.52	0.86
Majority vote2025.11		0.8	0.52	0.84
Debate with images2025.11		0.8	0.5	0.86
Debate about images2025.11		0.799	0.46	0.86
Direct prompt2025.11		0.788	0.48	0.85
Majority vote2025.11		0.775	0.45	0.84
Debate with images2025.11		0.773	0.49	0.83
Direct prompt2025.11		0.761	0.41	0.83
Debate with images2025.11		0.76	0.46	0.82
CoT prompt2025.11		0.759	0.45	0.78
Debate about images2025.11		0.745	0.42	0.81
Debate about images2025.11		0.735	0.43	0.79
Debate about images2025.11		0.72	0.43	0.77
Majority vote2025.11		0.692	0.4	0.74
Direct prompt2025.11		0.656	0.35	0.7
CoT prompt2025.11		0.63	0.33	0.66
Direct prompt2025.11		0.615	0.3	0.65
Majority vote2025.11		0.59	0.27	0.62
CoT prompt2025.11		0.473	0.16	0.42