Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LLM-MRD: LLM-Guided Multi-View Reasoning Distillation for Fake News Detection

Weilin Zhou, Shanwen Tan, Enhao Gu, Yurong Qian

Abstract

Multimodal fake news detection is crucial for mitigating societal disinformation. Existing approaches attempt to address this by fusing multimodal features or leveraging Large Language Models (LLMs) for advanced reasoning. However, these methods suffer from serious limitations, including a lack of comprehensive multi-view judgment and fusion, and prohibitive reasoning inefficiency due to the high computational costs of LLMs. To address these issues, we propose \textbf{LLM}-Guided \textbf{M}ulti-View \textbf{R}easoning \textbf{D}istillation for Fake News Detection ( \textbf{LLM-MRD}), a novel teacher-student framework. The Student Multi-view Reasoning module first constructs a comprehensive foundation from textual, visual, and cross-modal perspectives. Then, the Teacher Multi-view Reasoning module generates deep reasoning chains as rich supervision signals. Our core Calibration Distillation mechanism efficiently distills this complex reasoning-derived knowledge into the efficient student model. Experiments show LLM-MRD significantly outperforms state-of-the-art baselines. Notably, it demonstrates a comprehensive average improvement of 5.19\% in ACC and 6.33\% in F1-Fake when evaluated across all competing methods and datasets. Our code is available at https://github.com/Nasuro55/LLM-MRD

LLM-MRD: LLM-Guided Multi-View Reasoning Distillation for Fake News Detection

Abstract

Multimodal fake news detection is crucial for mitigating societal disinformation. Existing approaches attempt to address this by fusing multimodal features or leveraging Large Language Models (LLMs) for advanced reasoning. However, these methods suffer from serious limitations, including a lack of comprehensive multi-view judgment and fusion, and prohibitive reasoning inefficiency due to the high computational costs of LLMs. To address these issues, we propose \textbf{LLM}-Guided \textbf{M}ulti-View \textbf{R}easoning \textbf{D}istillation for Fake News Detection ( \textbf{LLM-MRD}), a novel teacher-student framework. The Student Multi-view Reasoning module first constructs a comprehensive foundation from textual, visual, and cross-modal perspectives. Then, the Teacher Multi-view Reasoning module generates deep reasoning chains as rich supervision signals. Our core Calibration Distillation mechanism efficiently distills this complex reasoning-derived knowledge into the efficient student model. Experiments show LLM-MRD significantly outperforms state-of-the-art baselines. Notably, it demonstrates a comprehensive average improvement of 5.19\% in ACC and 6.33\% in F1-Fake when evaluated across all competing methods and datasets. Our code is available at https://github.com/Nasuro55/LLM-MRD
Paper Structure (21 sections, 11 equations, 3 figures, 4 tables)

This paper contains 21 sections, 11 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. RELATED WORK
  3. Multimodal Fake News Detection
  4. Application of LLM in Fake News Detection
  5. Methodology
  6. Student Multi-view Reasoning
  7. Teacher Multi-view Reasoning
  8. Calibration Distillation
  9. Multi-View Fusion
  10. Loss Function
  11. Experiments
  12. Experimental Settings
  13. Overall Performance
  14. Ablation Study
  15. Parameter sensitivity analysis
  16. ...and 6 more sections

Figures (3)

  • Figure 2: LLM-MRD architecture overview. The student uses BERTdevlin2019bert, MAEhe2022masked, and CLIPradford2021learning to encode textual, visual, and cross-modal view features. The LLM teacher generates multi-view reasoning chains. Calibration Distillation aligns student features with the teacher's semantic space using projection layers before applying distillation losses. Calibrated features are integrated by Multi-View Fusion via cross-attention for prediction, while a Deep Supervision layer provides auxiliary guidance.
  • Figure 3: Analysis of hyperparameter sensitivity. This figure shows the impact of four different hyperparameters on the model's F1-real score across three datasets.
  • Figure 4: T-SNE visualization of test set features. Same color dots indicate the same label.