Metraipolis › archive › dc_f679796aaf26

A Parametric Linear Conic Projection for Structured Sparse Attention — arXiv:2611.04479

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04479
License (as archived)
CC0-1.0
Archived
2026-11-04T12:27:55.788Z
SHA-256 of the archived text
5983698d171b07253711beb152d7894943ef35869067962159551b9b82d7649d
Stable URL of this page
https://metraipolis.com/documento/dc_f679796aaf26

How to cite

A Parametric Linear Conic Projection for Structured Sparse Attention — arXiv:2611.04479 — arXiv (en), https://arxiv.org/abs/2611.04479 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f679796aaf26 (Metraipolis archive dc_f679796aaf26, sha256 5983698d171b07253711beb152d7894943ef35869067962159551b9b82d7649d)
BibTeX
@misc{dcf679796aaf26,
  title = {A Parametric Linear Conic Projection for Structured Sparse Attention — arXiv:2611.04479},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04479},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f679796aaf26; sha256 5983698d171b07253711beb152d7894943ef35869067962159551b9b82d7649d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04479. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In this paper, we introduce the Cone Alignment Index (CAI), a scale-invariant sparsity measure associated with a Lorentz-cone geometry. We provide a probabilistic analysis of the distribution of the CAI score, including a general theorem and explicit corollaries for Gaussian and uniform distributions. As a consequence, we show that, in the high-sparsity regime, the probability that the samples lie within the convex cone tends to one. Building upon this formulation, we derive two complementary projections: the \emph{Parametric Linear Conic Projection} (PLCP), an efficient closed-form approximation, and the \emph{Parametric Euclidean Conic Projection} (PECP), the exact Euclidean projection onto the cone boundary. Our analysis establishes several fundamental theoretical properties. We prove that PLCP and PECP are collinear, differ only by a positive radial scaling factor, and induce exactly the same sparsity threshold, thereby preserving an identical active support. Furthermore, we derive a support-identification theorem showing that, for any prescribed sparsity level $l<n$, the number of coefficients discarded at each iteration is theoretically controlled. Building on this theoretical foundation, we develop a bilevel PLCP framework for structured matrix sparsification. The proposed method naturally generates hardware-friendly structured sparsity patterns, including row-wise, column-wise, and diagonal masks, making it particularly well suited for accelerating Transformer attention mechanisms.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.04479 — 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 2026-11-04T12:26:47.158Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).