Metraipolis › archive › dc_57b86600f2db

Event-Driven Language Models with Sparse Neural Activity for Neuromorphic Hardware — arXiv:2608.30439

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30439
License (as archived)
CC0-1.0
Archived
2026-10-17T08:39:59.744Z
SHA-256 of the archived text
521dee53406e4438e63f2cfe00c43b28c2fc4876b479b941fbe3a30fb5e9bf67
Stable URL of this page
https://metraipolis.com/documento/dc_57b86600f2db

How to cite

Event-Driven Language Models with Sparse Neural Activity for Neuromorphic Hardware — arXiv:2608.30439 — arXiv (en), https://arxiv.org/abs/2608.30439 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_57b86600f2db (Metraipolis archive dc_57b86600f2db, sha256 521dee53406e4438e63f2cfe00c43b28c2fc4876b479b941fbe3a30fb5e9bf67)
BibTeX
@misc{dc57b86600f2db,
  title = {Event-Driven Language Models with Sparse Neural Activity for Neuromorphic Hardware — arXiv:2608.30439},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30439},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_57b86600f2db; sha256 521dee53406e4438e63f2cfe00c43b28c2fc4876b479b941fbe3a30fb5e9bf67},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30439. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Inference with transformer-based large language models (LLMs) is often limited by the memory-bound KV cache and quadratic attention cost. State-space models (SSMs) mitigate this through linear attention and fixed-size recurrent states, but their large dense linear projections remain computationally expensive even after quantization. We introduce a method that induces sparse neural activity in heavily quantized linear-attention models with minimal performance loss. Activations below a per-projection trainable threshold ($\pm \Delta$) are nullified while preserving crucial outliers, achieving comparable performance to dense models with up to 4$\times$ fewer effective arithmetic operations. Targeting a multi-core, multi-chip neuromorphic platform, where event-driven execution converts unstructured sparsity into throughput at both the compute and communication levels, a capability GPU architectures fundamentally lack, we project up to 37$\times$ higher throughput and 16$\times$ lower power versus edge GPU inference of a comparable transformer-based model, and up to 5.4$\times$ improvements over the non-sparsified baseline. These results position sparse, quantized linear-attention models as a natural fit for deploying LLMs on event-driven multi-core platforms.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.LG).
Categorie arXiv: Neural and Evolutionary Computing (cs.NE) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.30439 — 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 2026-10-17T08:13:26.941Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).