Author name cluster

Rui Xu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

28 papers

2 author rows

AAAI Conference 2026 Conference Paper

MCMoE: Completing Missing Modalities with Mixture of Experts for Incomplete Multimodal Action Quality Assessment

Huangbiao Xu
Huanqi Wu
Xiao Ke
Junyi Wu
Rui Xu
Jinglin Xu

Multimodal Action Quality Assessment (AQA) has recently emerged as a promising paradigm. By leveraging complementary information across shared contextual cues, it enhances the discriminative evaluation of subtle intra-class variations in highly similar action sequences. However, partial modalities are frequently unavailable at the inference stage in reality. The absence of any modality often renders existing multimodal models inoperable. Furthermore, it triggers catastrophic performance degradation due to interruptions in cross-modal interactions. To address this issue, we propose a novel Missing Completion Framework with Mixture of Experts (MCMoE) that unifies unimodal and joint representation learning in single-stage training. Specifically, we propose an adaptive gated modality generator that dynamically fuses available information to reconstruct missing modalities. We then design modality experts to learn unimodal knowledge and dynamically mix the knowledge of all experts to extract cross-modal joint representations. With a mixture of experts, missing modalities are further refined and complemented. Finally, in the training phase, we mine the complete multimodal features and unimodal expert knowledge to guide modality generation and generation-based joint representation extraction. Extensive experiments demonstrate that our MCMoE achieves state-of-the-art results in both complete and incomplete multimodal learning on three public AQA benchmarks.

PDF Details DOI

EAAI Journal 2026 Journal Article

Multi-reservoir echo state network based on the I Ching for accurate energy prediction

Bowen Wang
Jinghan He
Jizhen Liu
Rui Xu

Aiming at the problems of lack of theoretical guidance in structural design and limited feature extraction capability in traditional echo state network (ESN) for temporal modeling, this paper proposes an I Ching-based multi-reservoir ESN (IC-MRESN). The model employs a taiji reservoir for global regulation to capture long-term dependencies, a five-elements reservoir for local dynamic feature enhancement based on mutual generation and restriction relationships, and an eight diagrams reservoir for high-level feature decomposition and abstraction, thereby forming a synergistic and efficient hierarchical feature extraction mechanism. Adaptive optimization of information flow is achieved through dynamic interactions driven by energy intensity and state correlations among the reservoirs. Theoretical analysis demonstrates that the architecture enables efficient feature compression and transmission based on the information bottleneck principle, while exhibiting superior function approximation capability with an ultra-compact neuron scale. Finally, experimental validation on real-world energy data shows that the model outperforms other comparative models in prediction accuracy, generalization ability, and noise resistance under a simplified network topology.