# SAFformer:Improving Spiking Transformer via Active Predictive Filtering

arXiv: https://arxiv.org/abs/2605.08270 · HTML: https://mlsanity.com/paper/2605.08270/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 43 references that are in the map.
- First posted on arXiv 8 May 2026.

## Builds on (most cited first, 24 of 24)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [Spikformer: When Spiking Neural Network Meets Transformer](https://mlsanity.com/paper/2209.15425/) (2022)
- [Training Spiking Neural Networks Using Lessons From Deep Learning](https://mlsanity.com/paper/2109.12894/) (2021)
- [Spike-driven Transformer](https://mlsanity.com/paper/2307.01694/) (2023)
- [Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks](https://mlsanity.com/paper/2303.03667/) (2023)
- [Spike-driven Transformer V2: Meta Spiking Neural Network Architecture Inspiring the Design of Next-generation Neuromorphic Chips](https://mlsanity.com/paper/2404.03663/) (2024)
- [Integer-Valued Training and Spike-Driven Inference Spiking Neural Network for High-performance and Energy-efficient Object Detection](https://mlsanity.com/paper/2407.20708/) (2024)
- [Scaling Spike-driven Transformer with Efficient Spike Firing Approximation Training](https://mlsanity.com/paper/2411.16061/) (2024)
- [Speck: A Smart event-based Vision Sensor with a low latency 327K Neuron Convolutional Neuronal Network Processing Pipeline](https://mlsanity.com/paper/2304.06793/) (2023)
- [QKFormer: Hierarchical Spiking Transformer using Q-K Attention](https://mlsanity.com/paper/2403.16552/) (2024)
- [Masked Spiking Transformer](https://mlsanity.com/paper/2210.01208/) (2022)
- [Ternary Spike: Learning Ternary Spikes for Spiking Neural Networks](https://mlsanity.com/paper/2312.06372/) (2023)
- [Multi-scale Attention Network for Single Image Super-Resolution](https://mlsanity.com/paper/2209.14145/) (2022)
- [Gated Attention Coding for Training High-performance and Efficient Spiking Neural Networks](https://mlsanity.com/paper/2308.06582/) (2023)
- [Real Spike: Learning Real-valued Spikes for Spiking Neural Networks](https://mlsanity.com/paper/2210.06686/) (2022)
- [Quantized Spike-driven Transformer](https://mlsanity.com/paper/2501.13492/) (2025)
- [Spiking Transformer:Introducing Accurate Addition-Only Spiking Self-Attention for Transformer](https://mlsanity.com/paper/2503.00226/) (2025)
- [STAA-SNN: Spatial-Temporal Attention Aggregator for Spiking Neural Networks](https://mlsanity.com/paper/2503.02689/) (2025)
- [Spikingformer: A Key Foundation Model for Spiking Neural Networks](https://mlsanity.com/paper/2304.11954/) (2023)
- [FSTA-SNN:Frequency-based Spatial-Temporal Attention Module for Spiking Neural Networks](https://mlsanity.com/paper/2501.14744/) (2024)
- [SpiLiFormer: Enhancing Spiking Transformers with Lateral Inhibition](https://mlsanity.com/paper/2503.15986/) (2025)
- [MSVIT: Improving Spiking Vision Transformer Using Multi-scale Attention Fusion](https://mlsanity.com/paper/2505.14719/) (2025)

## Abstract (arXiv)

Spiking Neural Networks (SNNs) offer notable advantages in biological plausibility and energy efficiency, making them promising candidates for building low-power Transformers. However, existing Spiking Transformers largely adhere to a passive reactive paradigm, which struggles to focus on task-relevant information and incurs substantial computational overhead when processing redundant visual data. To overcome this fundamental yet underexplored limitation, we propose SAFformer, a novel Spiking Transformer architecture based on an active predictive filtering paradigm. Inspired by the brain's predictive coding mechanism, SAFformer actively suppresses predictable signals and focuses on salient visual features. Extensive experiments show that SAFformer establishes new state-of-the-art performance on CIFAR-10/100 and CIFAR10-DVS. Remarkably, on ImageNet-1K, it achieves 80.44% Top-1 accuracy with only 26.58M parameters and an energy consumption of 5.88 mJ, demonstrating an exceptional balance between accuracy and efficiency.