Metraipolis › archive › dc_dda248255db3
Lapis: Laplacian Spiking Attention via First-Spike Timing and Membrane Leakage — arXiv:2608.11865
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11865
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:52:32.161Z
- SHA-256 of the archived text
33570411d29fb8dbd0af2cf198df97c4b03df3a744e24b2781a27b33cbad9487- Stable URL of this page
https://metraipolis.com/documento/dc_dda248255db3
How to cite
Lapis: Laplacian Spiking Attention via First-Spike Timing and Membrane Leakage — arXiv:2608.11865 — arXiv (en), https://arxiv.org/abs/2608.11865 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dda248255db3 (Metraipolis archive dc_dda248255db3, sha256 33570411d29fb8dbd0af2cf198df97c4b03df3a744e24b2781a27b33cbad9487)
BibTeX
@misc{dcdda248255db3,
title = {Lapis: Laplacian Spiking Attention via First-Spike Timing and Membrane Leakage — arXiv:2608.11865},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11865},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dda248255db3; sha256 33570411d29fb8dbd0af2cf198df97c4b03df3a744e24b2781a27b33cbad9487},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11865. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Self-attention has become central to spiking vision transformers, yet its query-key scoring is still largely inherited from dense networks. Existing spiking variants either simplify dot product scoring or replace it with discrete operators, but spike timing, the native variable of a spiking network, does not directly define how tokens are related. We propose Lapis, a spiking attention mechanism that scores each token pair by the L1 distance between its query and key first-spike latency vectors under time-to-first-spike coding, and maps this distance to an affinity through a Laplacian kernel. The kernel's exponential decay matches the impulse response of a leaky integrate-and-fire membrane, so the accumulated latency difference determines the decay of a membrane trace, while row normalization reduces to a bit shift under power-of-two rounding. Scoring therefore needs only subtraction, absolute value, and accumulation, and removes all multiplication between query and key channels. Under a matched backbone and training schedule, Lapis reaches 96.56% top-1 accuracy on CIFAR-10, within 0.53 points of dot-product scoring. On ImageNet-1K, it reduces the estimated arithmetic energy of the attention path by 14.5x relative to dense dot-product attention. The deployed 6-bit model attains 83.25% top-1 accuracy at an estimated arithmetic energy of 3.28mJ per image. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.NE). Categorie arXiv: Neural and Evolutionary Computing (cs.NE). Pagina del lavoro: https://arxiv.org/abs/2608.11865 — 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 2026-10-19T05:48:07.355Z dalla pagina https://arxiv.org/catchup/cs.NE/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).