Metraipolis › archive › dc_b189479ced78

URCHIN: A Horizontal Spiking Language Model for Data-Constrained Pretraining — arXiv:2609.13899

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.13899
License (as archived): CC0-1.0
Archived: 2026-10-15T23:16:54.543Z
SHA-256 of the archived text: 043dcd1cb013de77f39626087630ce55d33829b2aedc93a7d0b3cae0fe692dfd
Stable URL of this page: https://metraipolis.com/documento/dc_b189479ced78

How to cite

URCHIN: A Horizontal Spiking Language Model for Data-Constrained Pretraining — arXiv:2609.13899 — arXiv (en), https://arxiv.org/abs/2609.13899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b189479ced78 (Metraipolis archive dc_b189479ced78, sha256 043dcd1cb013de77f39626087630ce55d33829b2aedc93a7d0b3cae0fe692dfd)

BibTeX

@misc{dcb189479ced78,
  title = {URCHIN: A Horizontal Spiking Language Model for Data-Constrained Pretraining — arXiv:2609.13899},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13899},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b189479ced78; sha256 043dcd1cb013de77f39626087630ce55d33829b2aedc93a7d0b3cae0fe692dfd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The BabyLM challenge measures how much language a model can learn from developmentally-plausible, child-scale data rather than internet-scale corpora, yet prior language models forgo the biological constraints of the neural circuitry that acquires human language: spiking neurons separated into excitatory and inhibitory populations wired by a recurrent lateral connectome. This paper presents URCHIN (Unified Recurrent Connectome with Horizontal Integrate-and-fire Neurons), which applies the Parallelized Hierarchical Connectome Spiking State-space Model (PHCSSM) to language modeling: leaky integrate-and-fire neurons coupled by a Dale's-law lateral connectome resolve each token through a multi-transmission loop that recirculates activity to a fixed point. The instantiation is deliberately minimal: a single horizontal layer of 128 neurons, no attention, and 4.23M parameters. Two implementations share one set of weights and produce identical benchmark scores, so URCHIN is trained once and deployed either way with no conversion step: a parallel state-space model (SSM) scan that is GPU-efficient for training, or an event-driven recurrent spiking neural network (RSNN) with constant-cost inference for CPU or neuromorphic edge deployment. Across all three BabyLM tracks (Strict-100M, Strict-Small, and Multilingual), URCHIN offers a biologically plausible, efficient, and directly deployable reference point.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL).
Categorie arXiv: Neurons and Cognition (q-bio.NC) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.13899 — 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 2026-10-15T22:51:10.012Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).