AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

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

Xiang Shi, Rui Zhang, Yu Liu, Yinpeng Liu, Zhu Liang, Qikai Cheng, Wei Ping Lu

TPAMI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add cross-modal alignment (ablation)Ours · CREMA-Daccuracy: 84.41 -> 84.95Improved
add memory unit (ablation)Ours · CREMA-Daccuracy: 64.25 -> 84.95Improved
add memory inheritance (ablation)Ours · CREMA-Daccuracy: 78.36 -> 84.95Improved
add text modality to audio (ablation)Ours · IEMOCAPaccuracy: 72.18 -> 75.07Improved
add visual modality to audio and text (ablation)Ours · IEMOCAPaccuracy: 72.18 -> 75.07Improved
replace fixed threshold with learnable gate (ablation)Ours · CREMA-Daccuracy: 84.95 -> 84.81No clear effect
replace fixed threshold with learnable gate (ablation)Ours · Kinetics-400accuracy: 79.35 -> 80.21Improved
weak-to-strong order instead of strong-to-weak or random (ablation)Ours · four benchmarksaccuracyImproved
vary initial modality order (ablation)Ours · four benchmarksaccuracy: less than 1% fluctuationNo clear effect
alternating EDM-guided training vs MLAOurs vs MLA · Kinetics-400accuracy: 72.66 -> 79.35Improved
EDM-guided method vs MLA under missing modalityOurs vs MLA · IEMOCAP 10% missingaccuracy: 52.57 -> 69.56Improved
alternating training vs joint concatOurs vs Concat · CREMA-Daccuracy: 70.30 -> 84.95Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy