A paper list of spiking neural networks, including papers, codes, and related websites. 本仓库收集脉冲神经网络相关的顶会顶刊以及CNS论文和代码，正在持续更新中。

A brain-inspired version of generative replay for continual learning with deep neural networks (e.g., class-incremental learning on CIFAR-100; PyTorch code).

Offical code of "QKFormer: Hierarchical Spiking Transformer using Q-K Attention" (NeurIPS 2024)

Spikingformer: A Key Foundation Model for Spiking Neural Networks (AAAI 2026)

The official PyTorch code for ICLR'22 Paper "Learning Fast, Learning Slow: A General Continual Learning Method based on Complementary Learning System""

Official PyTorch implementation of paper "A Hybrid Compact Neural Architecture for Visual Place Recognition" by M. Chancán (RA-L & ICRA 2020) https://doi.org/10.1109/LRA.2020.2967324

Enhancing the Performance of Transformer-based Spiking Neural Networks by SNN-optimized Downsampling with Precise Gradient Backpropagation

Spiking Global-Local Fusion Transformer

Implementation for the paper "SpaceNet: Make Free Space For Continual Learning" in PyTorch.

Multi-Task Structural Learning using Local Task Similarity induced Neuron Creation and Removal

Simple_Visual_Odometry_V1_01

This project implements a neuromorphic NoC using RISC-V nodes with custom hardware to simulate spiking neural networks. Tested on an FPGA, it supports up to 1,024 neurons and uses dedicated FIFO buffers for stable, low-power operation. The design combines RISC-V flexibility with hardware acceleration for efficient small-scale SNN applications.

This project implements a neuromorphic Network-on-Chip using customized RV32IMF RISC-V cores, a 2D mesh NoC, and neuron bank hardware to accelerate spiking neural networks. It supports event-driven spike communication via custom ISA extensions and AXI interfaces, and is implemented and tested on an FPGA.

This project implements a neuromorphic architecture using parallel RISC-V processing elements to model spiking neurons, either individually or virtually grouped. It will also develop encoder and decoder modules to benchmark the system on classical machine learning tasks.