SOTA Deception Detection benchmarks and papers with code

Benchmarks

Dataset Name	SOTA Method	Metric
XCHECK 1.0 (test)	NVILA	F1 Score87.44	28	2mo ago
DeceptionBench (full 600-sample)	DECOR	Response AUROC97.4	15	14d ago
MuDD Audio modality (test)	GPD	F1 Score20.1	11	2mo ago
MuDD Video modality (test)	GPD	F1 Score16.7	11	2mo ago
DOLOs	DecepGPT	Accuracy73.23	10	2mo ago
SEUMLD	GenLie	F1 Score42.39	10	2mo ago
Real-Life Trial		F1 Score93.55	10	2mo ago
MDPE	GenLie	F1 Score38.97	10	2mo ago
MDPE and DDEP	CLB-HBB-Bai-Rel-DDEP	Accuracy66.63	9	1mo ago
MU3D	DecepGPT	Accuracy61.25	9	2mo ago
Bag-of-Lies (BoL)	DecepGPT	Accuracy63.46	9	2mo ago
DeceptionBench	DECOR	Response AUROC93.5	6	14d ago
D-RepE (test)	Linear Probe	AUROC1	4	6d ago
DeceptArena (test)	Hybrid	False Assertion Score0.927	4	2mo ago
Liars' Bench Harm-Pressure Knowledge (test)	5-Layer Ensemble	AUROC0.91	3	1mo ago
Liars' Bench Insider Trading (test)	5-Layer Ensemble	AUROC0.953	3	1mo ago
Liars' Bench Convincing Game (test)		AUROC1	3	1mo ago
Liars' Bench Harm-Pressure Choice (test)		AUROC0.949	3	1mo ago
Liars' Bench Instructed Deception (test)		AUROC0.939	3	1mo ago
DOLOs, Bag-of-Lies, and MU3D Average	DecepGPT	Accuracy60.11	3	2mo ago
MU3D (test)	DecepGPT	Accuracy57.24	3	2mo ago
Bag-of-Lies (test)	DecepGPT	Accuracy59.62	3	2mo ago
DOLOs (test)	DecepGPT	Accuracy63.46	3	2mo ago
Deception Detection (test)	Heatmap	Behavioral Value (Acc Diff %)10	3	3mo ago
CharXiv modified images removed	gpt-5-thinking	Deception Rate9	2	3mo ago

Showing 25 of 27 rows