Adaptive Fission: Post-training Encoding for Low-latency Spike Neural Networks

Feng Chen (Microsoft AI / Stanford University) · Yihan Li (SUN YAT-SEN UNIVERSITY) · Yizhou Jiang (Tsinghua University, Tsinghua University) · Yuqian Liu (Tsinghua University, Tsinghua University) · Haichuan Gao (Tsinghua University, Tsinghua University) · Tianren Zhang (Tsinghua University, Tsinghua University) · Ying Fang (Xi'an Jiaotong University)

accuracy preservationadaptive fissionhigh-precision inferencehigh-sensitivity neuronslatency reduction neuromorphic hardwareneuron-specific precisionpopulation codingpost-training encodingpower consumptionpretrained snn architecturesrate codingspatial overheadspiking neural networksthreshold allocation

Spiking Neural Networks (SNNs) often rely on rate coding, where high-precision inference depends on long time-steps, leading to significant latency and energy cost—especially for ANN-to-SNN conversions. To address this, we propose Adaptive Fission, a post-training encoding technique that selectively splits high-sensitivity neurons into groups with varying scales and weights. This enables neuron-specific, on-demand precision and threshold allocation while introducing minimal spatial overhead. As a generalized form of population coding, it seamlessly applies to a wide range of pretrained SNN architectures without requiring additional training or fine-tuning. Experiments on neuromorphic hardware demonstrate up to 80\% reductions in latency and power consumption without degrading accuracy.