Metraipolis › archive › dc_e4dfa7dc43fe
METRO: Metric-Enhanced Token Routing Operator — arXiv:2610.05100
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.05100
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:30:55.166Z
- SHA-256 of the archived text
bf8e7ee43d969967940ee310637d774b3023634b096a2cd1f780ab1d5b733855- Stable URL of this page
https://metraipolis.com/documento/dc_e4dfa7dc43fe
How to cite
METRO: Metric-Enhanced Token Routing Operator — arXiv:2610.05100 — arXiv, https://arxiv.org/abs/2610.05100 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e4dfa7dc43fe (Metraipolis archive dc_e4dfa7dc43fe, sha256 bf8e7ee43d969967940ee310637d774b3023634b096a2cd1f780ab1d5b733855)
BibTeX
@misc{dce4dfa7dc43fe,
title = {METRO: Metric-Enhanced Token Routing Operator — arXiv:2610.05100},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.05100},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e4dfa7dc43fe; sha256 bf8e7ee43d969967940ee310637d774b3023634b096a2cd1f780ab1d5b733855},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05100. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
State-of-the-art neural operators scale to complex meshes via slice-and-process architectures, yet many rely on linear compatibility scores for latent tokenization. Under common feature normalization, such scores are equivalent to isotropic Euclidean clustering, while without normalization they induce unbounded linear decision regions. In both cases, they lack slice-specific anisotropic locality, which can lead to redundant and entangled latent slices. To address this, we propose Metric-Enhanced Token Routing Operator (METRO), a geometry-aware routing mechanism that replaces linear projection with a learnable Mahalanobis metric. By enabling each latent slice to learn a local anisotropic tensor, METRO shapes receptive fields into exponentially localized, oriented ellipsoids that naturally align with flow features like boundary layers and wakes. As a drop-in replacement, METRO yields consistent improvements across both Transformer and Mamba backbones. Empirically, our method achieves substantial performance gains on irregular domains, outperforming baselines on both standard PDE benchmarks and complex industrial design tasks. Finally, METRO exhibits enhanced robustness in out-of-distribution regimes across varying Reynolds numbers and geometric configurations. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.05100 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T05:27:05.765Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).