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

📰 ArXiv cs.AI

DecepGPT detects deception using schema-driven multimodal learning with multicultural datasets

advanced Published 26 Mar 2026

Action Steps

Collect and annotate multicultural datasets with intermediate reasoning cues
Develop schema-driven multimodal learning models to analyze audiovisual cues
Train and fine-tune DecepGPT on the collected datasets to improve generalization across domains and cultural contexts
Evaluate DecepGPT on benchmark datasets to assess its performance and reliability

Who Needs to Know This

AI engineers and researchers on a team can benefit from DecepGPT to improve deception detection in high-stakes settings, while data scientists can utilize the multicultural datasets for more robust model training

Key Insight

💡 DecepGPT provides verifiable evidence connecting audiovisual cues to final decisions, enabling more reliable deception detection

Key Takeaways

DecepGPT detects deception using schema-driven multimodal learning with multicultural datasets

Full Article

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

Abstract:
arXiv:2603.23916v1 Announce Type: cross 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 scen

Read full paper → ← Back to Reads