DAAD: Dynamic Analysis and Adaptive Discriminator for Fake News Detection

2024-08-20

arXiv.org

AI Summary

To address the rapid propagation of fake news in online social networks and the limitations of existing multimodal detection methods—namely, heavy reliance on manual feedback and poor generalizability—this paper proposes the Dynamic Analysis and Adaptive Discriminator (DAAD) framework. Methodologically, (1) it introduces a novel large language model (LLM)-based prompt self-reflection optimization mechanism grounded in Monte Carlo Tree Search (MCTS), enhancing knowledge-guided dynamism; (2) it constructs a soft-routing-driven scalable discriminator ensemble that explicitly models four deception patterns: emotional exaggeration, logical inconsistency, image manipulation, and semantic misalignment; and (3) it integrates knowledge-driven and semantic-driven dual inference pathways for multimodal joint reasoning. Evaluated on three real-world datasets, DAAD significantly outperforms state-of-the-art methods, demonstrating superior robustness and cross-domain generalizability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

In current web environment, fake news spreads rapidly across online social networks, posing serious threats to society. Existing multimodal fake news detection (MFND) methods can be classified into knowledge-based and semantic-based approaches. However, these methods are overly dependent on human expertise and feedback, lacking flexibility. To address this challenge, we propose a Dynamic Analysis and Adaptive Discriminator (DAAD) approach for fake news detection. For knowledge-based methods, we introduce the Monte Carlo Tree Search (MCTS) algorithm to leverage the self-reflective capabilities of large language models (LLMs) for prompt optimization, providing richer, domain-specific details and guidance to the LLMs, while enabling more flexible integration of LLM comment on news content. For semantic-based methods, we define four typical deceit patterns: emotional exaggeration, logical inconsistency, image manipulation, and semantic inconsistency, to reveal the mechanisms behind fake news creation. To detect these patterns, we carefully design four discriminators and expand them in depth and breadth, using the soft-routing mechanism to explore optimal detection models. Experimental results on three real-world datasets demonstrate the superiority of our approach. The code will be available at: https://github.com/SuXinqi/DAAD.

Problem

• Detects fake news using dynamic analysis and adaptive methods.
• Improves knowledge-based detection with Monte Carlo Tree Search.
• Identifies deceit patterns like emotional exaggeration and image manipulation.

Innovation

• Monte Carlo Tree Search for LLM prompt optimization
• Four deceit patterns defined for semantic analysis
• Soft-routing mechanism for optimal detection models