Bridging Thoughts and Words: Graph-Based Intent-Semantic Joint Learning for Fake News Detection

2025-09-01

AI Summary

Existing fake news detection methods over-rely on superficial semantic cues (e.g., sentiment words, stylistic features), resulting in limited generalizability and poor robustness under dynamic environments. To address this, we propose integrating news intent modeling into the detection framework—introducing the first joint learning of intent and semantics. Specifically, we construct a heterogeneous graph that explicitly encodes news entities, semantic representations, and multi-granular intents (coarse-grained dissemination intent and fine-grained deception strategies). We further design an entity-guided long-range contextual interaction mechanism and a dynamic path alignment module atop graph neural networks to establish deep intent–semantic correlations. Extensive experiments on four benchmark datasets demonstrate significant improvements over state-of-the-art methods, validating our model’s enhanced understanding of the intrinsic nature of news deception, as well as its superior generalizability and robustness in dynamic scenarios.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Application Domains: Misinformation & Fake News
• Machine Learning: Graph-based Machine Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Fake news detection is an important and challenging task for defending online information integrity. Existing state-of-the-art approaches typically extract news semantic clues, such as writing patterns that include emotional words, stylistic features, etc. However, detectors tuned solely to such semantic clues can easily fall into surface detection patterns, which can shift rapidly in dynamic environments, leading to limited performance in the evolving news landscape. To address this issue, this paper investigates a novel perspective by incorporating news intent into fake news detection, bridging intents and semantics together. The core insight is that by considering news intents, one can deeply understand the inherent thoughts behind news deception, rather than the surface patterns within words alone. To achieve this goal, we propose Graph-based Intent-Semantic Joint Modeling (InSide) for fake news detection, which models deception clues from both semantic and intent signals via graph-based joint learning. Specifically, InSide reformulates news semantic and intent signals into heterogeneous graph structures, enabling long-range context interaction through entity guidance and capturing both holistic and implementation-level intent via coarse-to-fine intent modeling. To achieve better alignment between semantics and intents, we further develop a dynamic pathway-based graph alignment strategy for effective message passing and aggregation across these signals by establishing a common space. Extensive experiments on four benchmark datasets demonstrate the superiority of the proposed InSide compared to state-of-the-art methods.

Problem

• Detecting fake news by combining intent and semantic signals
• Addressing limitations of surface-level semantic pattern detection
• Modeling deception clues through graph-based joint learning

Innovation

• Graph-based joint learning model
• Heterogeneous graph structure integration
• Dynamic pathway-based graph alignment