AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Unpacking the Implicit Norm Dynamics of Sharpness-Aware Minimization in Tensorized Models

Tianxiao Cao, Kyohei Atarashi, Hisashi Kashima

AAAI 2026 · Optimization › Implicit bias and generalization · approach: Sharpness-aware minimization theory · 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
replace SGD with SAMTR-ResNet-32 · CIFAR-10 80% label noisetest accuracy: 57.33 -> 67.13Improved
replace SGD with DASTR-ResNet-32 · CIFAR-10 80% label noisetest accuracy: 57.33 -> 60.74Improved
replace SGD with DASTR-ResNet-32 · CIFAR-10 60% label noisetest accuracy: 77.41 -> 77.73No clear effect
replace ADAM with SAMTucker · COVID tensor completionR2 score: 0.9482 -> 0.9485No clear effect
replace ADAM with DASTucker · COVID tensor completionR2 score: 0.9482 -> 0.9484No clear effect
replace SGD with DASTT-ResNet-18 · ImageNet-1k fine-tuningTop-1: 65.47 -> 66.16Improved
replace SGD with DASTT-ResNet-18 · ImageNet-1k fine-tuningruntime (s): 0.254 -> 0.254No clear effect
replace SAM with DASLoRETTA on OPT-6.7B · SuperGLUE/generation low-dataruntime: 2× -> 1.04×Improved
replace ADAM with DASFLoRA on RoBERTa-large · GLUE low-data avgaverage score: 83.78 -> 83.95Improved
replace ADAM with SAMFLoRA on RoBERTa-large · GLUE low-data avgaverage score: 83.78 -> 83.83No clear effect
replace ADAM with SAMLoRETTA on OPT-6.7B · SuperGLUE/generation low-dataaverage score: 70.25 -> 71.72Improved
replace ADAM with DASLoRETTA on OPT-6.7B · SuperGLUE/generation low-dataaverage score: 70.25 -> 71.41Improved

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