LGMay 26, 2025

Modality Equilibrium Matters: Minor-Modality-Aware Adaptive Alternating for Cross-Modal Memory Enhancement

arXiv:2506.00030v11 citationsh-index: 14
Originality Incremental advance
AI Analysis

This addresses the problem of biased learning in multimodal AI systems, especially under incomplete modality conditions, offering a new paradigm for optimizing training dynamics.

The paper tackles modality imbalance in multimodal fusion by proposing a Shapley-guided alternating training framework that adaptively prioritizes minor modalities, achieving state-of-the-art results on four benchmark datasets with improved balance and accuracy.

Multimodal fusion is susceptible to modality imbalance, where dominant modalities overshadow weak ones, easily leading to biased learning and suboptimal fusion, especially for incomplete modality conditions. To address this problem, we propose a Shapley-guided alternating training framework that adaptively prioritizes minor modalities to balance and thus enhance the fusion. Our method leverages Shapley Value-based scheduling to improve the training sequence adaptively, ensuring that under-optimized modalities receive sufficient learning. Additionally, we introduce the memory module to refine and inherit modality-specific representations with a cross-modal mapping mechanism to align features at both the feature and sample levels. To further validate the adaptability of the proposed approach, the encoder module empirically adopts both conventional and LLM-based backbones. With building up a novel multimodal equilibrium metric, namely, equilibrium deviation metric (EDM), we evaluate the performance in both balance and accuracy across four multimodal benchmark datasets, where our method achieves state-of-the-art (SOTA) results. Meanwhile, robustness analysis under missing modalities highlights its strong generalization capabilities. Accordingly, our findings reveal the untapped potential of alternating training, demonstrating that strategic modality prioritization fundamentally balances and promotes multimodal learning, offering a new paradigm for optimizing multimodal training dynamics.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes