A multimodal recommendation system (MRS), which leverages rich multimodal information to model user preferences, has recently attracted significant research interest. Most existing MRSs focus primarily on developing sophisticated encoders for feature extraction, typically relying on simple aggregation of interaction-based features for final predictions. However, this conventional paradigm fails to account for the critical semantic difference between high- and low-rating interactions: while high ratings indicate user preference, low ratings explicitly convey dissatisfaction. Such oversight of negative feedback semantics may significantly limit the system’s recommendation performance. Recently, sign graphs—which model positive and negative feedback signals separately—have gained considerable attention. Inspired by this approach, we propose Sign-Aware Multimodal Graph Recommendation (SiMGR), a novel framework incorporating signed graphs into multimodal recommendation systems. SiMGR fuses multimodal features with signed interactions in a unified graph framework by integrating modality-specific representations and applying user-level thresholds to separate positive and negative subgraphs. A balanced pseudo-edge augmentation strategy is introduced to alleviate sparsity and enhance generalization. Experiments on three public multimodal recommendation datasets show that SiMGR outperforms state-of-the-art baselines, achieving an average 4.28% improvement in NDCG@20.

Recently, many supervised deep learning-based methods have been investigated for large-scale MIMO channel estimation. However, the supervised methods require a large amount of ground truth channel as labeled data, which are difficult to obtain in practice. In this paper, we propose to learn the large-scale MIMO channel from compressed noisy measurements without any ground truth channel by introducing a loss function based on Generalized Stein’s Unbiased Risk Estimate (GSURE), which is an unbiased estimate of the projected mean-squared error (PMSE). Furthermore, we adopt deep equilibrium model (DEQ) to directly learn the equilibrium point of an iterative algorithm through an infinite-depth network, therefore it inherits the convergence performance of the underlying algorithm. We show that when using DEQ networks, employing GSURE for unsupervised learning can achieve performance close to that of supervised learning using MSE. Experiments demonstrate that the proposed GSURE-based unsupervised DEQ showcases superior channel estimation performance compared to various baselines when ground-truth channel is unavailable.