AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Sparse Fine-Tuning of Transformers for Generative Tasks

Wei Chen, Jingxi Yu, Zichen Miao, Qiang Qiu

ICCV 2025 · Architectures and efficiency › Pruning and efficient tuning · approach: Adapter-based parameter-efficient tuning · 2 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
adapt atoms instead of coefficients in toy signal reconstruction (ablation)attention block · toy high-frequency signal taskqualitative reconstructionImproved
tune only coefficients with atoms fixed (ablation)attention block · toy high-frequency signal taskqualitative reconstructionHurt
add sparse coefficients to attention output (ablation)attention block · toy signal reconstructionatoms needed: 100 -> 9 atomsImproved
use sparse dictionary fine-tuning instead of LoRAPixArt-Sigma · DreamBooth customizationtext alignment: 0.218 -> 0.235Improved
use sparse dictionary fine-tuning instead of LoRAPixArt-Sigma · DreamBooth customizationdiversity (Vendi): 4.707 -> 5.205Improved
use rho=0.02 instead of LoRA for ImageRewardPixArt-Sigma · DreamBooth customizationImageReward: 0.220 -> 0.182Hurt
increase coefficient density rho from 0.02 to 0.04 (ablation)sparse fine-tuning · DreamBooth customizationImageReward: 0.182 -> 0.431Improved
increase dictionary size M from 16 to 2048 (ablation)sparse fine-tuning · DreamBooth customization, rho=0.01fidelity: 0.547 -> 0.692Improved
increase dictionary size M from 16 to 2048 (ablation)sparse fine-tuning · DreamBooth customization, rho=0.01diversity (Vendi): 5.867 -> 3.525Hurt
increase sparsity (lower density rho) (ablation)sparse fine-tuning · DreamBooth customizationfidelityMixed
remove unimportant atoms by raising sparsity at edit timefine-tuned PixArt-Sigma · image editingqualitative text alignment of editImproved
sparse atom adaptation versus rank-1 LoRA under modified prompt (ablation)PixArt-Sigma · image editingstabilityImproved

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