GenLie: A Global-Enhanced Lie Detection Network under Sparsity and Semantic Interference

2026-03-14

AI Summary

Video-based deception detection faces significant challenges due to the sparsity and transience of deceptive cues, as well as interference from identity-related noise, which hinders the learning of discriminative representations. To address these issues, this work proposes GenLie, a novel network featuring a local–global collaborative modeling architecture. The local component captures subtle deceptive signals through fine-grained feature extraction, while a global supervision mechanism suppresses identity-induced noise and enhances representation robustness. This dual strategy effectively mitigates semantic interference and data sparsity. Extensive experiments demonstrate that GenLie consistently outperforms state-of-the-art methods across three public datasets encompassing both high- and low-stakes deception scenarios, validating its effectiveness and strong generalization capability.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Video-based lie detection aims to identify deceptive behaviors from visual cues. Despite recent progress, its core challenge lies in learning sparse yet discriminative representations. Deceptive signals are typically subtle and short-lived, easily overwhelmed by redundant information, while individual and contextual variations introduce strong identity-related noise. To address this issue, we propose GenLie, a Global-Enhanced Lie Detection Network that performs local feature modeling under global supervision. Specifically, sparse and subtle deceptive cues are captured at the local level, while global supervision and optimization ensure robust and discriminative representations by suppressing identity-related noise. Experiments on three public datasets, covering both high- and low-stakes scenarios, show that GenLie consistently outperforms state-of-the-art methods. Source code is available at https://github.com/AliasDictusZ1/GenLie.

Problem

• lie detection
• sparsity
• semantic interference
• deceptive behavior
• visual cues

Innovation

• Global-Enhanced Lie Detection
• Sparse Deceptive Cues
• Identity-Related Noise Suppression
• Video-Based Lie Detection
• Local-Global Feature Modeling