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

TL;DR

GenLie is a novel video-based lie detection network that captures subtle deceptive cues locally while using global supervision to enhance robustness against noise, outperforming existing methods across multiple datasets.

Contribution

The paper introduces GenLie, a global-enhanced network that effectively models sparse deceptive cues with global supervision, improving lie detection accuracy.

Findings

Outperforms state-of-the-art methods on three public datasets

Effective in both high- and low-stakes scenarios

Robustly suppresses identity-related noise

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.