AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Generalized Threshold Optimization with Harmony Multi-Threshold Neurons for Accurate ANN-to-SNN Conversion

Wenhan Zhang, Zihan Huang, Tong Bu, Tiejun Huang, Zhaofei Yu

AAAI 2026 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add communication mechanism to H-MT 4/Opt (ablation)H-MT 4/Opt · ImageNet ViT-base (T=16)accuracy (%): 69.214 -> 74.942Improved
add communication mechanism to H-MT 4/999 (ablation)H-MT 4/999 · ImageNet ViT-base (T=16)accuracy (%): 63.482 -> 73.878Improved
use generalized threshold optimization instead of 99.9% quantile, no communication (ablation)H-MT 4 · ImageNet ViT-base (T=16)accuracy (%): 63.482 -> 69.214Improved
use generalized threshold optimization instead of 99.9% quantile (ablation)H-MT 4 · ImageNet ResNet34 (T=16)accuracy (%): 72.85 -> 75.80Improved
use generalized threshold optimization instead of 99.9% quantile (ablation)H-MT 4 · ImageNet VGG-16bnaccuracy (%): 72.19 -> 71.94 (T=4); 72.15 -> 72.92 (T=16)Mixed
use generalized threshold optimization instead of 99.9% quantile (ablation)H-MT 8 · ImageNet ResNet34 (T=2)accuracy (%): 72.908 -> 76.276Improved
use generalized threshold optimization instead of 99.9% quantile (ablation)H-MT 8 · ImageNet ResNet34 (T=2)energy ratio: 0.5435 -> 0.2986Improved
double the 99.9% quantile thresholdH-MT 8/999 · ImageNet ResNet34 (T=2)accuracy (%): 72.908 -> 75.702Improved
increase thresholds from n=3 to n=4H-MT /Opt · ImageNet ResNet34 (T=16)accuracy (%): 73.91 -> 75.80Improved
increase time-steps from 2 to 16H-MT 4/Opt · ImageNet ResNet34accuracy (%): 71.73 -> 75.80Improved
replace identity mapping with H-MT (y=x) neuronsH-MT 4/Opt · ImageNet ResNet18 (T=16)accuracy (%): 70.95 -> 67.61Hurt
increase thresholds from n=4 to n=8 at T=2H-MT /Opt · ImageNet ResNet34accuracy (%): 71.73 -> 76.276Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy