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

Weilin Zhou; Shanwen Tan; Enhao Gu; Yurong Qian

cs.CLMar 10, 2026

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

Authors: Weilin Zhou, Shanwen Tan, Enhao Gu, Yurong Qian

Organizations: Xinjiang University, Urumqi, China · Joint International Research Laboratory of Silk Road Multilingual Cognitive Computing, Xinjiang University, Urumqi, China · Sichuan University, Chengdu, China

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 LLM-Guided Multi-View Reasoning Distillation for Fake News Detection ( 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

Figures & tables

Panel of Figure 1 — Figure 1: Comparison of our proposed ”teacher-student” framework with previous methods. (a) Traditional fusion methods (e.g., Attention, Add) lack deep contextual understanding and are easily misled by manipulations like image-text mismatching, leading to misjudgments. (b) Our approach employs an LLM teacher for deep multi-view reasoning and distills this knowledge to a efficient student model, enabling it to learn robust features for accurate classification.

Figure 2: LLM-MRD architecture overview. The student uses BERT [ 35 ] , MAE [ 34 ] , and CLIP [ 36 ] 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.

		Weibo				Weibo-21				GossipCop
Category	Method	Acc	F1-Fake	F1-Real	AUC	Acc	F1-Fake	F1-Real	AUC	Acc	F1-Fake	F1-Real	AUC
Multimodal multi-domain methods	EANN [ 28 ]	0.827	0.829	0.825	0.873	0.870	0.862	0.875	0.894	0.864	0.594	0.920	0.852
	FND-CLIP [ 29 ]	0.907	0.908	0.907	0.953	0.943	0.940	0.946	0.962	0.880	0.638	0.928	0.871
	MIMoE-FND [ 30 ]	0.928	0.928	0.928	0.972	0.956	0.955	0.957	0.977	0.895	0.698	0.938	0.879
	KEN [ 31 ]	0.935	0.935	0.934	0.967	0.935	0.937	0.932	0.971	0.881	0.646	0.928	0.873
Multimodal single-domain methods	RaCMC [ 24 ]	0.915	0.917	0.914	0.921	0.942	0.938	0.943	0.962	0.879	0.641	0.927	0.838

Table 1: Comparison of LLM-MRD with the latest multi-domain fake news detection methods from Weibo, Weibo-21, and GossipCop.

Model	Weibo			Weibo-21			GossipCop
Model	Acc	F1-Fake	F1-Real	Acc	F1-Fake	F1-Real	Acc	F1-Fake	F1-Real
LLM-MRD	0.953	0.954	0.952	0.959	0.957	0.968	0.902	0.719	0.946
Iterative Calibration distillation
- w/o $\mathcal{L}_{\text{text}}$	0.929 $\downarrow$ 2.5%	0.924 $\downarrow$ 3.1%	0.920 $\downarrow$ 3.4%	0.937 $\downarrow$ 2.3%	0.933 $\downarrow$ 2.5%	0.938 $\downarrow$ 3.1%	0.881 $\downarrow$ 2.3%	0.672 $\downarrow$ 6.5%	0.915 $\downarrow$ 3.3%
- w/o $\mathcal{L}_{\text{image}}$	0.925 $\downarrow$ 2.9%	0.918 $\downarrow$ 3.8%	0.922 $\downarrow$ 3.2%	0.923 $\downarrow$ 3.8%	0.935 $\downarrow$ 2.3%	0.937 $\downarrow$ 3.2%	0.874 $\downarrow$ 3.1%	0.674 $\downarrow$ 6.3%	0.913 $\downarrow$ 3.5%
- w/o $\mathcal{L}_{\text{cross}}$	0.917 $\downarrow$ 3.8%	0.907 $\downarrow$ 4.9%	0.912 $\downarrow$ 4.2%	0.945 $\downarrow$ 1.5%	0.921 $\downarrow$ 3.8%	0.927 $\downarrow$ 4.2%	0.889 $\downarrow$ 1.4%	0.665 $\downarrow$ 7.5%	0.901 $\downarrow$ 4.8%

Table 2: Ablation study results showing performance drops for different model variants.

Panel of Figure 3 — 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.

Panel of Table 3 — Table 3: LLM-MRD explainability case study on two challenging examples from the Weibo21 test set.

Figure 4: T-SNE visualization of test set features. Same color dots indicate the same label.

Method	Params	Reason latency	Time Complexity
MIMoE-FND	397M	20.5ms	$O(k\cdot n_{e}\cdot(n^{2d}))$
KEN	382M	19.7ms	$O(L\cdot(n^{2d}))$
RaCMC	424M	18.8ms	$O(L\cdot(n^{2d}))$
GLPN-LLM	7.07B	158ms	$O(L_{\text{llm}}\cdot n_{\text{llm}}^{2d})$
LLM-MRD	358M	17.2ms	$O(L\cdot(n^{d}))$

Table 4: Comparative Reasoning efficiency analysis of LLM-MRD and baseline models

Fake News Detection

Fake news detection identifies false or misleading claims by analyzing content, context, evidence, and how information spreads.

See all 93 papers

Retrieval-Augmented Multimodal Model for Fake News Detection
Apr 20, 2026Yiheng Li, Weihai Lu, Hanyi Yu +1Fake News Detection Text Classification
University of International Business and Economics · Beijing, China · Peking University +4
Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality
Apr 8, 2026Kai Qian, Weijie Shi, Jiaqi Wang +7Fake News Detection Multimodal Robustness
Soochow University, China · Hong Kong University of Science and Technology, China · Computer Science and Engineering, The Chinese University of Hong Kong, China +5
MRAFnd: Multimodal Retrieval-Augmented Framework for Zero-Shot Fake News Detection
Aug 2, 2026Lehan Zhang, Yinlei Cheng, Shiqi Hu Yiheng Zhou +2Social Media Analysis Fake News Detection
Beijing Institute of Fashion Technology, Beijing, China · Shenyang University of Technology, Shenyang, China

Apr 20, 2026J/K move · Enter open · S save

Retrieval-Augmented Multimodal Model for Fake News Detection

Yiheng Li, Weihai Lu, Hanyi Yu, Yue Wang

University of International Business and Economics · Beijing, China · Peking University+4

Fake News Detection Text Classification Multimodal Reasoning

Open paper AskPDF arXiv

In recent years, multimodal multidomain fake news detection has garnered increasing attention. Nevertheless, this direction presents two significant challenges: (1) Failure to Capture Cross-Instance Narrative Consistency: existing models usually evaluate each news in isolation, fail to capture cross-instance narrative consistency, and thus struggle to address the spread of cluster based fake news driven by social media; (2) Lack of Domain Specific Knowledge for Reasoning: conventional models, which rely solely on knowledge encoded in their parameters during training, struggle to generalize to new or data-scarce domains (e.g., emerging events or niche topics). To tackle these challenges, we introduce Retrieval-Augmented Multimodal Model for Fake News Detection (RAMM). First, RAMM employs a Multimodal Large Language Model (MLLM) as its backbone to capture cross-modal semantic information from news samples. Second, RAMM incorporates an Abstract Narrative Alignment Module. This component adaptively extracts abstract narrative consistency from diverse instances across distinct domains, aggregates relevant knowledge, and thereby enables the modeling of high-level narrative information. Finally, RAMM introduces a Semantic Representation Alignment Module, which aligns the model's decision-making paradigm with that of humans - specifically, it shifts the model's reasoning process from direct inference on multimodal features to an instance-based analogical reasoning process. Extensive experimental results on three public datasets validate the efficacy of our proposed approach. Our code is available at the following link: https://github.com/li-yiheng/RAMM

Abstract

Fake News Detection

Explore similar work

Retrieval-Augmented Multimodal Model for Fake News Detection

Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality

MRAFnd: Multimodal Retrieval-Augmented Framework for Zero-Shot Fake News Detection