DecepGPT: Schema-Driven Deception Detection with Multicultural Datasets and Robust Multimodal Learning

[Submitted on 25 Mar 2026 (v1), last revised 7 Jul 2026 (this version, v4)]

View PDF HTML (experimental)

Abstract:Multimodal deception detection aims to identify deceptive behavior by analyzing audiovisual cues for forensics and security. In these high-stakes settings, investigators need verifiable evidence connecting audiovisual cues to final decisions, along with reliable generalization across domains and cultural contexts. However, existing benchmarks provide only binary labels without intermediate reasoning cues. Datasets are also small with limited scenario coverage, leading to shortcut learning. We address these issues through three contributions. First, we construct reasoning datasets by augmenting existing benchmarks with structured cue-level descriptions and reasoning chains, enabling models to output auditable reports. Second, we release T4-Deception, a multicultural dataset based on the unified “To Tell the Truth” television format implemented across four countries. With 1695 samples, it is the largest non-laboratory deception detection dataset. Third, we propose two modules for robust learning under small-data conditions. Stabilized Individuality-Commonality Synergy (SICS) refines multimodal representations by combining learnable global priors with sample-adaptive residuals and applying polarity-aware recalibration. Distilled Modality Consistency (DMC) aligns modality-specific predictions with the fused multimodal predictions via knowledge distillation to prevent unimodal shortcut learning. Experiments on three established benchmarks and our novel dataset demonstrate that our method achieves state-of-the-art performance in both in-domain and cross-domain scenarios, while exhibiting superior transferability across diverse cultural contexts. The datasets and code are available at this link.

Submission history

From: Jiajian Huang Mr. [view email]
[v1] Wed, 25 Mar 2026 04:06:36 UTC (2,638 KB)
[v2] Sat, 11 Apr 2026 11:12:51 UTC (2,866 KB)
[v3] Sun, 7 Jun 2026 07:53:12 UTC (2,865 KB)
[v4] Tue, 7 Jul 2026 12:17:42 UTC (3,151 KB)