← Search

Desong Zhang

2 accepted papers

2026

Robustify Spiking Neural Networks via Dominant Singular Deflation under Heterogeneous Training Vulnerability

ICLR 2026poster

Spiking Neural Networks (SNNs) process information via discrete spikes, enabling them to operate at remarkably low energy levels. However, our experimental observations reveal a striking vulnerability when SNNs are trained using the mainstream method—direct encoding combined with backpropagation thr…

Cited by 0SourcecodeScholar