Metraipolis › archive › dc_eb563e354715

Attention Is All You Need (to Avoid Spurious Oscillations) — arXiv:2609.13531

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13531
License (as archived)
CC0-1.0
Archived
2026-10-15T22:48:34.825Z
SHA-256 of the archived text
df81e4771df9f2a16262587a9265c7e65aa405cd163673a7381921bcea8061bb
Stable URL of this page
https://metraipolis.com/documento/dc_eb563e354715

How to cite

Attention Is All You Need (to Avoid Spurious Oscillations) — arXiv:2609.13531 — arXiv (en), https://arxiv.org/abs/2609.13531 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eb563e354715 (Metraipolis archive dc_eb563e354715, sha256 df81e4771df9f2a16262587a9265c7e65aa405cd163673a7381921bcea8061bb)
BibTeX
@misc{dceb563e354715,
  title = {Attention Is All You Need (to Avoid Spurious Oscillations) — arXiv:2609.13531},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13531},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eb563e354715; sha256 df81e4771df9f2a16262587a9265c7e65aa405cd163673a7381921bcea8061bb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13531. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Can attention move a shock across several cells in one update without breaking it? We develop a conservative, fixed grid finite-volume scheme in which a CFL-conditioned attention flux selects upstream information according to the transport required by the current time step. One-dimensional inviscid Burgers transport is used as the central mechanism test: the same learned flux remains reliable in the conventional small-step regime and, with a time step four times larger, preserves sharp shocks while using one stage per update. A standard fifth-order WENO scheme with third-order strong-stability-preserving Runge-Kutta time integration (WENO-5+SSP-RK3) is included alongside controlled Forward Euler comparisons to separate flux selection from time integration. The learned attention shifts upstream with the local transport reach and becomes more selective near shocks; inference-time interventions and retrained ablations show that transport-scale information and state-dependent selection contribute directly to performance. Directional two-dimensional scalar Burgers transport and the one-dimensional shallow-water system then test whether the conservation-scale-selection principle transfers beyond the original scalar setting. The results support attention as a learnable information stencil for conservative large-step shock transport, while identifying finite candidate reach and problem-dependent robustness as the present limits.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Fluid Dynamics (physics.flu-dyn).
Pagina del lavoro: https://arxiv.org/abs/2609.13531 — 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 2026-10-15T22:29:03.516Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).