FADRW: A Feature-Aware Modulated and Dynamically Reweighted Loss for Few-Shot Linguistic Steganalysis

2026-06-03

AI Summary

This work addresses the challenges of extreme class imbalance and the masking of steganographic signals by benign textual content in few-shot linguistic steganalysis on social media. To tackle these issues, the authors propose FADRW, a novel loss function framework that jointly mitigates decision bias and feature marginalization at the optimization level. FADRW employs a dynamic reweighting mechanism to adaptively adjust sample weights and incorporates a feature-aware modulation module to reshape the feature space, thereby enhancing the discriminability of subtle steganographic signals. Experimental results on three real-world social media datasets demonstrate that FADRW significantly outperforms existing methods, achieving substantial improvements in detection performance, particularly in few-shot scenarios.

Abstract

The ubiquity of social media platforms facilitates malicious linguistic steganography, posing significant security risks. However, detection is severely hampered by two fundamental issues during model training. Firstly, extreme class imbalance (less than 1% steganographic samples) induces a strong decision bias. Secondly, the invisibility of generative steganography means its features are nearly indistinguishable from benign text; this similarity, compounded by their extreme rarity, leads to severe feature marginalization, where faint steganographic signals are completely overwhelmed. To directly address these optimization-level challenges, we propose FADRW (Feature-Aware Modulated and Dynamically Reweighted Loss), a novel loss function framework engineered for few-shot steganalysis. FADRW employs Dynamic Reweighting to progressively counteract decision bias, and a Feature-Aware Modulation module to structurally reshape the feature space, preventing feature marginalization by enhancing the separability of these subtle features. Extensive experiments on datasets from three real-world social platforms demonstrate that FADRW significantly outperforms state-of-the-art methods, particularly in the challenging few-shot steganographic sample scenario.

Problem

• few-shot
• linguistic steganalysis
• class imbalance
• feature marginalization
• generative steganography

Innovation

• Few-shot steganalysis
• Dynamic reweighting
• Feature-aware modulation
• Class imbalance
• Linguistic steganography