AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Towards Training-Free and Accurate ANN-to-SNN Conversion via Activation-Aware Redistribution

Honglin Cao, Shuai Wang, Zijian Zhou, Ammar Belatreche, Wenjie Wei, Yu Liang, Yu Yang, Rui Xi, Malu Zhang, Haizhou Li

AAAI 2026 · Architectures and efficiency › Spiking networks · approach: Spiking transformers · 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 channel-wise threshold (ablation)layer-wise IF baseline, ViT-L · ImageNet-1K T=64top-1 accuracy: 73.17 -> 84.27Improved
add channel-wise threshold (ablation)layer-wise IF baseline, ViT-L · ImageNet-1K T=8top-1 accuracy: 0.42 -> 53.15Improved
add channel-wise offset (ablation)layer-wise IF baseline, ViT-L · ImageNet-1K T=64top-1 accuracy: 73.17 -> 84.49Improved
add channel-wise offset (ablation)layer-wise IF baseline, ViT-L · ImageNet-1K T=8top-1 accuracy: 0.42 -> 66.34Improved
add channel-wise offset to channel-wise threshold (full AIF) (ablation)CW-threshold, ViT-L · ImageNet-1K T=4top-1 accuracy: 1.73 -> 58.18Improved
add channel-wise threshold to channel-wise offset (full AIF) (ablation)CW-offset, ViT-L · ImageNet-1K T=4top-1 accuracy: 0.17 -> 58.18Improved
add both components (full AIF) (ablation)baseline, ViT-L · ImageNet-1K T=32top-1 accuracy: 63.45 -> 85.33Improved
increase time steps from 32 to 64AIF, ViT-B/16 · CIFAR10 / CIFAR100accuracy: 98.7 -> 98.9; 91.8 -> 92.0Improved
increase time steps from 8 to 64AIF, BERT-base · GLUE averageaverage score: 77.1 -> 82.7Improved
AIF conversion vs direct-training SpikeLMAIF, BERT-base · GLUE averageaverage score: 76.5 -> 82.7Improved
AIF vs Spike-ZIP training-free conversionAIF ViT-L · ImageNet-1K T=64top-1 accuracy: 83.82 -> 85.52Improved

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