AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Noise Stability of Transformer Models

Themistoklis Haris, Zihan Zhang, Yuichi Yoshida

ICLR 2026 · Architectures and efficiency › Attention and transformers · approach: Transformer expressivity and learnability · 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 noise stability regularizationdecoder-only Transformer · Modular additioniterations to emergence: 4500 -> 3300 (36% acceleration)Improved
add noise stability regularizationdecoder-only Transformer · Noisy k-sparse paritytraining iterations to grokking: ~35% speed-upImproved
add noise stability regularizationdecoder-only Transformer · WikiText-2 next-token predictioniterations to high validation accuracy: ~75% fewer iterationsImproved
use noise stability bound instead of average sensitivity boundspectral tail mass prediction · GPT-2, BERT, RoBERTa, Gemmatail mass bound: RoBERTa 0.19 -> 0.02; GPT-2 0.003 -> 0.0005Improved

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