Keywords
Modality Risk
Expert Routing
Missing Data
Calibration
Clinical Safety
Abstract
Research spanning trustworthy multimodal biomedical AI and biomedical signal representation increasingly joins methods that were developed for different objects and decisions. Here, trust-informed conditional routing among modality experts in biomedical classification is compared with a coupled transformer autoencoder for separating multi-region neural latent dynamics to determine which claims can travel across those boundaries and which remain context dependent. The review draws on two focal records and 12 established sources already present in the project evidence cache. Its comparative framework links modality risk, expert routing, and missing data to downstream questions of calibration and clinical safety. Comparison reveals recurring trade-offs among modality risk, expert routing, and missing data. These trade-offs do not support a universal ranking; instead, they identify the operating envelope within which each method remains credible and the perturbations most likely to expose fragile conclusions. The article concludes with a research agenda built around transparent comparators, targeted stress tests, and evidence records that can be reused without overstating causal or practical reach.
References
Chang, Y., Cheng, A., Wu, C., Wang, Z., Chen, J., Chattopadhyay, T., ... & Bogdan, P. (2026). TIER-MoE: Trust-Informed Expert Routing via Conditional Modality Risk for Multimodal Fusion in Biomedical Classification. arXiv preprint arXiv:2607.27289.
Sristi, R. D., Narasimha, S., Huang, J., Despatin, A., Musall, S., Gilja, V., & Mishne, G. (2026, April). Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics. In International Conference on Learning Representations (Vol. 2026, pp. 71457-71488).
Chae, D. E., & Lee, S. P. (2025). Dual Routing Mixture-of-Experts for Multi-Scale Representation Learning in Multimodal Emotion Recognition. Electronics, 14(24), 4972. https://doi.org/10.3390/electronics14244972
Yang, L., Fan, W., & Bouguila, N. (2023). Deep Clustering Analysis via Dual Variational Autoencoder With Spherical Latent Embeddings. IEEE Transactions on Neural Networks and Learning Systems, 34(9), 6303-6312. https://doi.org/10.1109/tnnls.2021.3135460
Xiao, X., Wang, X., Qi, S., & Li, H. (2026). Uncertainty-aware mixture of experts for robust multimodal sentiment analysis. Pattern Recognition, 179, 113417. https://doi.org/10.1016/j.patcog.2026.113417
Pahič, R., Lončarević, Z., Gams, A., & Ude, A. (2021). Robot skill learning in latent space of a deep autoencoder neural network. Robotics and Autonomous Systems, 135, 103690. https://doi.org/10.1016/j.robot.2020.103690
Reddy Bora, S., & Kishan Anapu, S. (2025). Intelligent LLM Orchestration: Advanced Mixture of Experts Routing for Large Language Model Systems. International Journal of Science and Research (IJSR), 1833-1838. https://doi.org/10.21275/sr25627232328
Raynukaazhakarsamy, D. (2026). SC-VAEGAN: spectral-constrained variational autoencoder with generative adversarial networks for robust unsupervised deep clustering with density-aware latent representations. Journal of Artificial Intelligence Machine Learning and Neural Network, 127. https://doi.org/10.55529/jaimlnn.61.127.135
Abdullakutty, F., Akbari, Y., Al-Maadeed, S., Saady, R. A., & Rasheed, H. M. A. (2026). Attention-enhanced Mixture of Experts for multimodal breast cancer detection. Biomedical Signal Processing and Control, 118, 109576. https://doi.org/10.1016/j.bspc.2026.109576
Guo, J., Fan, W., Amayri, M., & Bouguila, N. (2025). Deep clustering analysis via variational autoencoder with Gamma mixture latent embeddings. Neural Networks, 183, 106979. https://doi.org/10.1016/j.neunet.2024.106979
Raza, W. H., Wen, Y., Shah, A. B., Shen, Y., Martinez-Lemus, J., Schiess, M. C., et al. (2026). NeuroMoE++: Patient-Adaptive Multi-Level Multimodal Fusion With Mixture-of-Experts for Neurological Disorder Classification. IEEE Transactions on Biomedical Engineering, 73(8), 2784-2794. https://doi.org/10.1109/tbme.2026.3691657
Talibart, H., & Gilis, D. (2026). Learning a chemistry-aware latent space for molecular encoding and generation with a large-scale transformer variational autoencoder. Journal of Cheminformatics, 18(1). https://doi.org/10.1186/s13321-026-01243-0
Alhaj, Z., & Altairi, A. (2026). Spatial–Spectral Operator Mixture Transformer With Mixture-of-Experts Fusion for Endoscopic Classification. IEEE Access, 14, 49928-49944. https://doi.org/10.1109/access.2026.3679531
Li, J., Zhang, B., & Zhang, D. (2018). Shared Autoencoder Gaussian Process Latent Variable Model for Visual Classification. IEEE Transactions on Neural Networks and Learning Systems, 29(9), 4272-4286. https://doi.org/10.1109/tnnls.2017.2761401
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Copyright (c) 2026 Preston Walsh, Reid Hart, Sawyer Snyder (Author)