Explainable Detection of Implicit Influential Patterns in Conversations via Data Augmentation

2025-06-17

AI Summary

In the digital age, malicious actors increasingly exploit implicit linguistic strategies embedded in dialogues to conduct psychological infiltration and covert manipulation—patterns that are non-explicit and highly context-dependent, rendering them difficult for existing models to detect. To address this challenge, we formulate implicit influence pattern identification as a fine-grained, dialogue-level multi-label classification task—the first such formulation in the literature. We propose an LLM-driven, interpretable data augmentation paradigm to mitigate severe annotation scarcity. Furthermore, we design a unified framework integrating dialogue-context awareness, multi-task learning, and an interpretable neural architecture. Experimental results demonstrate a 6% improvement in implicit influence detection accuracy, alongside 33% and 43% gains in F1-score for influence technique identification and victim vulnerability classification, respectively. Crucially, our approach significantly enhances both interpretability and practical utility of detection.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

In the era of digitalization, as individuals increasingly rely on digital platforms for communication and news consumption, various actors employ linguistic strategies to influence public perception. While models have become proficient at detecting explicit patterns, which typically appear in texts as single remarks referred to as utterances, such as social media posts, malicious actors have shifted toward utilizing implicit influential verbal patterns embedded within conversations. These verbal patterns aim to mentally penetrate the victim's mind in order to influence them, enabling the actor to obtain the desired information through implicit means. This paper presents an improved approach for detecting such implicit influential patterns. Furthermore, the proposed model is capable of identifying the specific locations of these influential elements within a conversation. To achieve this, the existing dataset was augmented using the reasoning capabilities of state-of-the-art language models. Our designed framework resulted in a 6% improvement in the detection of implicit influential patterns in conversations. Moreover, this approach improved the multi-label classification tasks related to both the techniques used for influence and the vulnerability of victims by 33% and 43%, respectively.

Problem

• Detect implicit influential patterns in digital conversations
• Identify locations of influential elements within dialogues
• Improve detection accuracy using data augmentation techniques

Innovation

• Detects implicit influential patterns via data augmentation
• Identifies influential elements within conversation locations
• Uses state-of-the-art language models for dataset enhancement