AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

S2FT: Parameter-Efficient Fine-Tuning in Sparse Spectrum Domain

Baoquan Zhang, Zhehao Yu, Lisai Zhang, Kenghong Lin, Tianran Chen, Yuxi Sun, Yunming Ye, Yao He

CVPR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Adapter-based parameter-efficient tuning · 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
use S2FT instead of FourierFT (half the parameters)ViT-B/16 · VTAB-1kmean accuracy: 72.8 -> 73.6Improved
use S2FT instead of FourierFT (same 0.16% parameters)ViT-B/16 · VTAB-1kmean accuracy: 72.8 -> 74.1Improved
use S2FT instead of FourierFTViT-B/16 · FGVCmean accuracy: 87.9 -> 89.1Improved
use S2FT instead of FourierFTRoBERTa-base · GLUEaverage score: 85.0 -> 85.6Improved
use S2FT instead of FourierFTRoBERTa-large · GLUEaverage score: 88.0 -> 88.3Improved
use S2FT instead of FourierFTLLaMA2-13B · MT-BenchGPT-4 score: 5.82 -> 5.89Improved
use S2FT instead of FourierFTLLaMA2-7B · MT-BenchGPT-4 score: 5.18 -> 5.21No clear effect
use spectrum-prior sampling instead of random sampling (ablation)S2FT · VTAB-1kmean accuracy: 72.0 -> 73.6Improved
use spectrum-prior sampling instead of low-frequency sampling (ablation)S2FT · VTAB-1kmean accuracy: 72.8 -> 73.6Improved
increase gamma from 0.5 to 1.5 (ablation)S2FT · VTAB-1kmean accuracy: 72.3 -> 73.6Improved
increase gamma from 1.5 to 5.0 (ablation)S2FT · VTAB-1kmean accuracy: 73.6 -> 72.7Hurt
pre-estimate weight change with full fine-tuning (5 epochs) instead of cumulative gradient (ablation)S2FT · VTAB-1kmean accuracy: 73.6 -> 73.8No clear effect

Later papers that compared with this method

1 later accepted paper reports S2FT as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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