# Latency and accuracy tradeoffs in Spiking Neural Networks

arXiv: https://arxiv.org/abs/2609.35260 · HTML: https://mlsanity.com/paper/2609.35260/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 21 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 17 of 17)

- [Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition](https://mlsanity.com/paper/1804.03209/) (2018)
- [The Heidelberg spiking datasets for the systematic evaluation of spiking neural networks](https://mlsanity.com/paper/1910.07407/) (2019)
- [Learning Delays in Spiking Neural Networks using Dilated Convolutions with Learnable Spacings](https://mlsanity.com/paper/2306.17670/) (2023)
- [ConSmax: Hardware-Friendly Alternative Softmax with Learnable Parameters](https://mlsanity.com/paper/2402.10930/) (2024)
- [Co-learning synaptic delays, weights and adaptation in spiking neural networks](https://mlsanity.com/paper/2311.16112/) (2023)
- [Reconsidering the Energy Efficiency of Spiking Neural Networks Inference from Analytical Perspectives](https://mlsanity.com/paper/2409.08290/) (2024)
- [Advancing Spatio-Temporal Processing in Spiking Neural Networks through Adaptation](https://mlsanity.com/paper/2408.07517/) (2024)
- [LMUFormer: Low Complexity Yet Powerful Spiking Model With Legendre Memory Units](https://mlsanity.com/paper/2402.04882/) (2024)
- [Sorbet: A Neuromorphic Hardware-Compatible Transformer-Based Spiking Language Model](https://mlsanity.com/paper/2409.15298/) (2024)
- [Otters: An Energy-Efficient SpikingTransformer via Optical Time-to-First-Spike Encoding](https://mlsanity.com/paper/2509.18968/) (2025)
- [Global-Local Convolution with Spiking Neural Networks for Energy-efficient Keyword Spotting](https://mlsanity.com/paper/2406.13179/) (2024)
- [Efficient Speech Command Recognition Leveraging Spiking Neural Network and Curriculum Learning-based Knowledge Distillation](https://mlsanity.com/paper/2412.12858/) (2024)
- [SpikCommander: A High-performance Spiking Transformer with Multi-view Learning for Efficient Speech Command Recognition](https://mlsanity.com/paper/2511.07883/) (2025)
- [BSO: Binary Spiking Online Optimization Algorithm](https://mlsanity.com/paper/2511.12502/) (2025)
- [Temporal Information Reconstruction and Non-Aligned Residual in Spiking Neural Networks for Speech Classification](https://mlsanity.com/paper/2501.00348/) (2024)
- [Congestion-Aware Dynamic Axonal Delay for Spiking Neural Networks](https://mlsanity.com/paper/2605.01291/) (2026)
- [Benchmarking spiking neural networks across sensing modalities on edge devices](https://mlsanity.com/paper/2609.00026/) (2026)

## Abstract (arXiv)

Spiking neural networks are attractive for low-power speech command recognition, yet their latency has received far less attention than their energy efficiency, and their multi-timestep execution is widely assumed to make them slower than quantized neural networks. This paper challenges the assumption that more local timesteps necessarily imply higher network latency. By overlapping computation across adjacent layers at the timestep level, SNNs may complete execution in less time than comparable bit-serial QNNs. However, this overlap relies on spikes firing on incomplete inputs, and a spike once generated cannot be withdrawn, so its error persists and reduces accuracy. Waiting for more input before firing would seem to improve accuracy at the cost of reduced overlap. Yet we find and prove that this intuition fails at some layers, where even a small increase in waiting can change spike timing and downstream computation, making the network both slower and less accurate. We therefore propose a Pipeline Delay Search method which selects each layer's delay by balancing task-level accuracy gains against added network latency. We then adapt the selected configurations through spike-based quantization-aware training and bounded tuning of firing thresholds and initial membrane potentials. Together, these steps form Falcon, a framework for Fine-grained Analysis of Latency and Controlled firing which systematically analyzes and optimizes SNN latency under a spatial analog compute-in-memory mapping with shared digital engines. We evaluate Falcon on GSCV2 and SSC, achieving competitive accuracies of 96.31 and 83.02 at modeled network-core latencies of 119.64 and 124.00us, respectively. Together, our analysis and results show that SNNs can compute more yet finish faster, and wait longer yet predict worse, highlighting why Falcon matters for both latency and accuracy.