AI Paper Atlas

Research map / Image generation / Editing and style transfer

SliderEdit: Continuous Image Editing with Fine-Grained Instruction Control

Arman Zarei, Samyadeep Basu, Mobina Pournemat, Sayan Nag, Ryan A. Rossi, Soheil Feizi

CVPR 2026 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 10 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
apply the adapter globally to all image and text tokens (GSTLoRA instead of STLoRA) (ablation)SliderEdit · single-instruction face editsContinuity (CLIP): 0.2538 -> 0.2998Improved
apply the adapter globally to all image and text tokens (GSTLoRA instead of STLoRA) (ablation)SliderEdit · single-instruction face editsDisentanglement (DINO): 0.1071 -> 0.1168Hurt
use STLoRA instead of explicit CFG for edit strengthFLUX-Kontext · single-instruction face editsContinuity (CLIP): 0.1993 -> 0.2538Improved
use GSTLoRA instead of Implicit CFGFLUX-Kontext · single-instruction face editsContinuity (CLIP): 0.1547 -> 0.2998Improved
use GSTLoRA instead of Explicit CFGFLUX-Kontext · single-instruction face editsExtrapolation (CLIP): 0.2731 -> 0.2648Hurt
use STLoRA instead of Explicit CFGFLUX-Kontext · single-instruction face editsDisentanglement (ID cosine distance): 0.3415 -> 0.2550Improved
use SliderEdit STLoRA instead of Concept SliderConcept Slider · single-instruction face editsDisentanglement (ID cosine distance): 0.7091 -> 0.2550Improved
use STLoRA instead of Continuous Attribute ControlContinuous Attribute Control · single-instruction face editsContinuity (CLIP): 0.1891 -> 0.2538Improved
swap base model from FLUX-Kontext to Qwen-Image-EditSliderEdit STLoRA · 3-instruction face editsContinuity average: 0.3691 -> 0.4345Improved
swap base model from FLUX-Kontext to Qwen-Image-EditSliderEdit STLoRA · 1-instruction face editsContinuity average: 0.2709 -> 0.2319Hurt
swap base model from FLUX-Kontext to Qwen-Image-EditSliderEdit STLoRA · 1-instruction face editsExtrapolation average: 0.2630 -> 0.3214Improved
swap base model from FLUX-Kontext to Qwen-Image-EditSliderEdit STLoRA · 3-instruction face editsDisentanglement average (lower better): 0.2762 -> 0.3630Hurt

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