Multimodal Counterfactual Fusion with Causality-Guided Contrastive Learning for Noise-Robust Recommendation
Open Access
- 13 October 2026
- journal article
- research article
- Published by MDPI AG in Future Internet
- Vol. 18 (10) , 551
Abstract
Multimodal Recommender Systems (MRSs) leverage heterogeneous data to alleviate data sparsity, yet they remain vulnerable to spurious correlations and modality noise that obscure the rationale behind recommendations. To address this, we propose the Multimodal Counterfactual Fusion with the Causality-Guided Contrastive Learning (MCFC) framework. MCFC integrates three core modules: a Multimodal Structural Causal Model that motivates the assumed causal paths of multimodal biases; a Counterfactual Feature Fusion module with causal effect-inspired weighting that estimates each modality’s contribution via score perturbations and suppresses noisy features; and a Causality-Guided Contrastive Learning mechanism that builds counterfactual in-batch negatives to enforce invariant and robust representations. Extensive experiments on four public Amazon datasets show that MCFC consistently outperforms state-of-the-art baselines on Recall@K and NDCG@K, improving Recall@20 by 29.9% over the strongest baseline on the highly sparse Baby dataset. Ablation studies confirm the contribution of each module, evidencing MCFC’s robustness to modality noise and strong adaptability in sparse interaction settings.Keywords
Funding Information
- Department of Education of Liaoning Province (LJ212510165008)
- Department of Education of Liaoning Province (LJ212413631012)
This publication has 40 references indexed in Scilit:
- SGA-GNN: Semantic-Guided Adaptive Graph Neural Network for Cold-Start Multimodal Recommendation2026
- Cross-modal local interest contrast with Dual-graph denoising for multimodal recommendationSignal, Image and Video Processing, 2026
- Backpropagation-Based Counterfactual Explanations for Quasi-Nonlinear Fuzzy Cognitive MapsIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2026
- Efficient Total Causal Effect Computation in Fuzzy Cognitive Maps for Scalable Explainable Artificial IntelligenceIEEE Systems, Man, and Cybernetics Letters, 2026
- TDIC: Time-Aware Disentanglement of Interest and Conformity in Mobile App Recommendations2025
- MGDCF: Distance Learning via Markov Graph Diffusion for Neural Collaborative FilteringIEEE Transactions on Knowledge and Data Engineering, 2024
- Information flow-based fuzzy cognitive maps with enhanced interpretabilityGranular Computing, 2023
- Causal Incremental Graph Convolution for Recommender System RetrainingIEEE Transactions on Neural Networks and Learning Systems, 2022
- MMCF2018
- Attentive Collaborative Filtering2017