AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

SpikingLM: Towards Fully Spiking Language Model

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

ICML 2026 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks

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 Spike2Max attention instead of spiking self-attentionSpikingLM backbone · Language modelingperplexity: 24.58 -> 19.78Improved
replace SSA with Spike2MaxSpikingLM backbone · Language modelingenergy (mJ): 1.580 -> 1.457Improved
add Distribution-aware Scaling alone (ablation)Base spiking BERT · Pre-trainingperplexity: 24.47Improved
add Spike2Max alone without scaling (ablation)Base spiking BERT · Pre-trainingperplexity: 256.39Hurt
add Spike2Max on top of Distribution-aware Scaling (ablation)SpikingLM · Pre-trainingperplexity: 24.47 -> 19.78Improved
use Distribution-aware Scaling instead of RMSNorm (ablation)SpikingLM · Pre-trainingperplexity: 22.18 -> 19.78Improved
use Distribution-aware Scaling instead of RMSNorm (ablation)SpikingLM · Pre-trainingenergy (mJ): 1.643 -> 1.545Improved
use learnable Distribution-aware Scaling instead of fixed scale 2 (ablation)SpikingLM · Pre-trainingperplexity: 45.93 -> 19.78Improved
use Spike2Max instead of standard Softmax (ablation)SpikingLM · Pre-trainingperplexity: 18.29 -> 19.78Hurt
use Spike2Max instead of standard Softmax (ablation)SpikingLM · Pre-trainingenergy (mJ): 1.761 -> 1.457Improved
add learnable temporal gamma to Max-Sub (Spike2Max vs Max-Sub) (ablation)SpikingLM · Pre-trainingperplexity: 22.75 -> 19.78Improved
use Spike2Max/DAS SpikingLM instead of SpikeLMGLUE · GLUEaverage score: 76.5 -> 77.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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