SOURCE-LINKED INTELLIGENCE

Large Language Models with At Most One Spike per Neuron

arXiv · AI, language, vision and robotics · article · Sep 4, 2026 · UTC

Leveraging their inherent sparse event-driven computation, spiking neural networks (SNNs) offer a promising path toward energy-efficient large language models (LLMs). Time-to-first-spike (TTFS) coding generates at most one spike per neuron within a time window, yielding extremely low firing rates. However, conventional TTFS SNNs are restricted to specific structures, making it challenging to encode certain blocks in LLM -- such as layer normalization and matrix multiplication --using TTFS. To overcome this limitation, we introduce a reference-based strategy specifically to encode the four core

Read original source ↗ Open in workspace

recordType

paper

region

Global