Metraipolis › archive › dc_f833d207347f
SCRAMJet: An Ultra Low Latency JetClass-Scale Transformer for the HL-LHC Level-1 Trigger — arXiv:2611.00086
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00086
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:24:41.516Z
- SHA-256 of the archived text
7eb4fd405324f8193150c2729530c38af4e2012185bd77c0f29e4809dd0fca6a- Stable URL of this page
https://metraipolis.com/documento/dc_f833d207347f
How to cite
SCRAMJet: An Ultra Low Latency JetClass-Scale Transformer for the HL-LHC Level-1 Trigger — arXiv:2611.00086 — arXiv (en), https://arxiv.org/abs/2611.00086 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f833d207347f (Metraipolis archive dc_f833d207347f, sha256 7eb4fd405324f8193150c2729530c38af4e2012185bd77c0f29e4809dd0fca6a)
BibTeX
@misc{dcf833d207347f,
title = {SCRAMJet: An Ultra Low Latency JetClass-Scale Transformer for the HL-LHC Level-1 Trigger — arXiv:2611.00086},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00086},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f833d207347f; sha256 7eb4fd405324f8193150c2729530c38af4e2012185bd77c0f29e4809dd0fca6a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00086. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Level-1 Triggers at the LHC classify collisions within a few microseconds on FPGAs, where a network is a fully unrolled circuit whose cost is area rather than time. We present SCRAMJet, a transformer encoder that replaces multi-head attention with a low-rank projection, trains under per-parameter learned fixed-point quantization, and compiles to Verilog with distributed arithmetic optimization. On the ten-class JetClass dataset it reaches $71.4 \pm 0.6\%$ accuracy with 0.157M multiply-accumulates, 6.1 accuracy points above the strongest published resource-constrained model at an eighth of the operations, and meets timing on a Virtex UltraScale+ VU13P at 147 ns using 25.7% of the logic. We also show that resource-aware training silently removes attention channels: one whose learned bit-width reaches zero still carries signal during training but is absent from the compiled circuit, so the loss never sees it go. We call these virtual channels, and a floor on the attention datalane that prevents them is worth 3.1 accuracy points at matched budget. Code and trained checkpoints are available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Instrumentation and Detectors (physics.ins-det) ; Machine Learning (cs.LG); High Energy Physics - Experiment (hep-ex). Pagina del lavoro: https://arxiv.org/abs/2611.00086 — 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 2026-11-03T11:02:37.753Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).