Research map / Architectures and efficiency / Mixture of experts
Guiding Mixture-of-Experts with Temporal Multimodal Interactions
Xing Han, Hsing-Huan Chung, Joydeep Ghosh, Paul Pu Liang, Suchi Saria
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use MERGE temporal RUS-guided routing instead of FuseMoE | MoE multimodal model · PAMAP2 | accuracy: 87.74 -> 91.37 | Improved |
| use MERGE instead of I2MoE | MoE multimodal model · MIMIC-IV IHM | AUROC: 83.28 -> 85.40 | Improved |
| use MERGE instead of FuseMoE | MoE multimodal model · MOSI | accuracy: 75.65 -> 72.04 | Hurt |
| use MERGE instead of FuseMoE | MoE multimodal model · MIMIC-IV LOS | F1: 75.18 -> 74.43 | Hurt |
| add redundancy auxiliary loss (ablation) | MERGE · PAMAP2 | accuracy | Improved |
| add uniqueness auxiliary loss (ablation) | MERGE · PAMAP2, MIMIC-IV, Opportunity | accuracy | Improved |
| add synergy auxiliary loss (ablation) | MERGE · MIMIC-IV IHM and Opportunity | accuracy | Improved |
| increase maximum temporal RUS time lag from 1 to 10 (ablation) | MERGE · PAMAP2, Opportunity, WESAD | accuracy | Improved |
| repeat temporal RUS segments to form longer sequences (ablation) | MERGE · PAMAP2, Opportunity, WESAD | accuracy | Improved |
| use multi-scale BATCH estimator instead of naive step-wise RUS (ablation) | MERGE · PAMAP2, WESAD, Opportunity | accuracy | No clear effect |
Closest papers by meaning
- $\texttt{I$^2$MoE}$: Interpretable Multimodal Interaction-aware Mixture-of-Experts (ICML 2025 · similarity 0.88)
- R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts (ICML 2025 · similarity 0.88)
- Routing Experts: Learning to Route Dynamic Experts in Existing Multi-modal Large Language Models (ICLR 2025 · similarity 0.88)
- MoDE: A Mixture-of-Experts Model with Mutual Distillation among the Experts (AAAI 2024 · similarity 0.87)
- FLAME: Adaptive Mixture-of-Experts for Continual Multimodal Multi-Task Learning (NeurIPS 2026 · similarity 0.87)
- HumanSense: From Multimodal Perception to Empathetic Context-Aware Responses Through Reasoning MLLMs (AAAI 2026 · similarity 0.86)
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.
