Metraipolis › archive › dc_5ae8f25df9ce
Rethinking the Tradeoff Between Temporal Encoding and Nonlinear Computation in Spiking Language Models — arXiv:2610.10933
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.10933
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T05:56:16.040Z
- SHA-256 of the archived text
d1fb5ecdc1d6292243592028cd453c33cff7fe389fde04926c4e1756d253382b- Stable URL of this page
https://metraipolis.com/documento/dc_5ae8f25df9ce
How to cite
Rethinking the Tradeoff Between Temporal Encoding and Nonlinear Computation in Spiking Language Models — arXiv:2610.10933 — arXiv (en), https://arxiv.org/abs/2610.10933 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5ae8f25df9ce (Metraipolis archive dc_5ae8f25df9ce, sha256 d1fb5ecdc1d6292243592028cd453c33cff7fe389fde04926c4e1756d253382b)
BibTeX
@misc{dc5ae8f25df9ce,
title = {Rethinking the Tradeoff Between Temporal Encoding and Nonlinear Computation in Spiking Language Models — arXiv:2610.10933},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.10933},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5ae8f25df9ce; sha256 d1fb5ecdc1d6292243592028cd453c33cff7fe389fde04926c4e1756d253382b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10933. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Spiking language models face a tradeoff between representing continuous semantic features over short temporal windows and retaining costly nonlinear attention operations. We introduce Spora, which jointly designs spike encodings and attention operators. Binary temporal weights let $T$ spikes represent compositional values with up to $T$ bits of capacity, compared with $O(\log_2 T)$ bits for spike-count readout. Unipolar Binary Spiking (UBS) uses thresholds and spike-triggered residual decay to produce non-negative integer codes; Bipolar Binary Spiking (BBS) separates sign and magnitude and learns a scale for signed activations. These representations support accumulation-and-shift dot products and integer-exponent mappings in attention. With four time steps, Spora achieves 76.6 average GLUE score and 44.1 CoLA MCC, improving over SpikeLM by 1.2 and 6.2 points, respectively. Extending BBS to six steps raises these scores to 78.2 and 47.4. Conditional-decay analysis, matched-budget activation-quantization comparisons, event-workload statistics, and fixed-point evaluation further characterize the connection between encoding fidelity and computational cost. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.10933 — 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 2026-10-09T05:47:01.400Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).