Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics: Causal Claims and Confounding

Causal language requires a design that separates the proposed mechanism from selection effects, omitted variables, and other plausible explanations. This structured evidence review evaluates "Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics" alongside nine author-disjoint, topically matched publications in multi-region neural dynamics. It compares construct definitions, evaluation choices, operating assumptions, and reported limitations instead of treating bibliographic similarity as empirical equivalence. Viewed through causal interpretation and confounding control, the map separates claims supported by the available record from questions that still require full-text extraction, replication, or new experiments. The synthesis is interpretive rather than meta-analytic and therefore does not present a pooled effect estimate or a new causal result. The resulting agenda requires a stated causal estimand, defensible controls, negative checks, and sensitivity analyses for unmeasured confounding.

1. Sristi, R. D., Narasimha, S. M., Huang, J., Despatin, A., Musall, S., Gilja, V., & Mishne, G. (2026). Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics. In The Fourteenth International Conference on Learning Representations.
2. Talibart, H., & Gilis, D. (2025). Learning a chemistry-aware latent space for molecular encoding and generation with a large-scale Transformer Variational Autoencoder. . https://doi.org/10.64898/2025.12.19.695394 DOI
3. Xin, Q., & Kass, R.-E. (2025). Identifying interactions across brain areas while accounting for individual-neuron dynamics with a Transformer-based variational autoencoder. Advances in Neural Information Processing Systems 38, 131831-131856. https://doi.org/10.52202/085713-3968 DOI
4. Čeperković, B. (2025). Transformer-Based Autoencoder for Denoising EEG and ECoG Signals: A Breakthrough in Neural Signal Processing. . https://doi.org/10.21203/rs.3.rs-6689702/v1 DOI
5. Laturnus, S., & Berens, P. (2021). MorphVAE: Generating Neural Morphologies from 3D-Walks using a Variational Autoencoder with Spherical Latent Space. . https://doi.org/10.1101/2021.06.14.448271 DOI
6. Sharma, A. (2026). Interruptible Neural Execution: Intra-Layer Epistemic Interrupts and Latent Fact Injection for Transformer Factual Grounding. . https://doi.org/10.2139/ssrn.6815160 DOI
7. Kim, H., Kong, K., Kim, J.-K., Lee, J., Cha, G., Jang, H.-D., Wee, D., & Kang, S.-J. (2023). Controlling 3d Human Action with Transformer Variational Autoencoder in Latent Space. . https://doi.org/10.2139/ssrn.4600274 DOI
8. Li, X., Zhao, Z., Song, D., Zhang, Y., Pan, J., Wu, L., Huo, J., Niu, C., & Wang, D. (2020). Latent Factor Decoding of Multi-Channel EEG for Emotion Recognition Through Autoencoder-Like Neural Networks. Frontiers in Neuroscience, 14. https://doi.org/10.3389/fnins.2020.00087 DOI
9. Saffari, F., Schmidt, M.-N., Buscema, M., & Bruni, L.-E. (2024). Emotion Recognition from Brain Connectivity Based on Multi-Head Attention Transformer and Autoencoder Network. . https://doi.org/10.36227/techrxiv.170420879.92145482/v1 DOI
10. Song, Z., Wu, S., Zhou, T., & Wang, Y. (2025). Extracting Preserved Neural Latent Dynamics Across Tasks using Convolutional Transformer-based Variational Autoendecoder*. 2025 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 1-4. https://doi.org/10.1109/embc58623.2025.11251780 DOI