Refinement Is Inherently Editable: Training-Free Prompt-to-Prompt Image Editing with Generative Refinement Network

Hugging Face Daily Papers Papers
image-editing prompt-to-prompt training-free diffusion-model generative-ai

Summary

RefineEdit is a training-free prompt-to-prompt image editing method that uses a generative refinement network to enhance edit localization and background preservation, achieving top benchmark scores.

Text-guided image editing must introduce the requested changes while preserving unrelated source content. Diffusion-based editors rely on spatial controls whose inaccuracies can leave edits incomplete or alter unrelated regions. Causal autoregressive editors face a further constraint: their fixed decoding order limits revision of earlier decisions. We introduce RefineEdit, a training-free prompt-to-prompt image editing framework built on a Generative Refinement Network. Our key idea is to couple edit localization with content generation through the global refinement of binary image codes, allowing editing evidence to be reassessed as the image evolves. RefineEdit initializes an editing branch from an intermediate source state, reusing the emerging layout. We compare the probabilities assigned by the two branches to the same source-sampled bits, using their signed differences to select editable positions and bits. Selected bits follow editing refinement, while the remaining bits copy the evolving source state. To stabilize editing across refinement steps, adaptive spatial freezing limits unnecessary mask expansion, while finite bit locking keeps recently selected bits editable. The framework requires no additional training, external masks, or attention control. Across nine editing categories of PIE-Bench, RefineEdit achieves the best background-preservation scores in PSNR, LPIPS, MSE and SSIM, together with the highest whole-image and edited-region CLIP scores among the evaluated methods.
Original Article
View Cached Full Text

Cached at: 09/21/26, 03:22 PM

Paper page - Refinement Is Inherently Editable: Training-Free Prompt-to-Prompt Image Editing with Generative Refinement Network

Source: https://huggingface.co/papers/2609.20633

Abstract

Text-guidedimageeditingmustintroducetherequestedchangeswhilepreservingunrelatedsourcecontent.Diffusion-basededitorsrelyonspatialcontrolswhoseinaccuraciescanleaveeditsincompleteoralterunrelatedregions.Causalautoregressiveeditorsfaceafurtherconstraint:theirfixeddecodingorderlimitsrevisionofearlierdecisions.WeintroduceRefineEdit,atraining-freeprompt-to-promptimageeditingframeworkbuiltonaGenerativeRefinementNetwork.Ourkeyideaistocoupleeditlocalizationwithcontentgenerationthroughtheglobalrefinementofbinaryimagecodes,allowingeditingevidencetobereassessedastheimageevolves.RefineEditinitializesaneditingbranchfromanintermediatesourcestate,reusingtheemerginglayout.Wecomparetheprobabilitiesassignedbythetwobranchestothesamesource-sampledbits,usingtheirsigneddifferencestoselecteditablepositionsandbits.Selectedbitsfolloweditingrefinement,whiletheremainingbitscopytheevolvingsourcestate.Tostabilizeeditingacrossrefinementsteps,adaptivespatialfreezinglimitsunnecessarymaskexpansion,whilefinitebitlockingkeepsrecentlyselectedbitseditable.Theframeworkrequiresnoadditionaltraining,externalmasks,orattentioncontrol.AcrossnineeditingcategoriesofPIE-Bench,RefineEditachievesthebestbackground-preservationscoresinPSNR,LPIPS,MSEandSSIM,togetherwiththehighestwhole-imageandedited-regionCLIPscoresamongtheevaluatedmethods.

View arXiv pageView PDFGitHub7Add to collection

Get this paper in your agent:

hf papers read 2609\.20633

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.20633 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.20633 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.20633 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

RefGC-SR^2: Reference-guided Generated Content Super-Resolution and Refinement

Hugging Face Daily Papers

This paper introduces a new task, reference-guided generated content super-resolution-refinement (RefGC-SR²), which simultaneously recovers high-resolution details and refines generative artifacts using a frequency-aware diffusion transformer model. The method leverages a high-resolution reference image to improve the quality of AI-generated images during post-processing.

Uni-Edit: Intelligent Editing Is A General Task For Unified Model Tuning

Hugging Face Daily Papers

Uni-Edit proposes using intelligent image editing as a single general task to simultaneously improve unified multimodal models' understanding, generation, and editing capabilities, with an automated data synthesis pipeline creating complex editing instructions.

UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing

arXiv cs.CL

UltraX proposes a function-calling refinement framework for large-scale pre-training data that introduces insertion alongside deletion and modification, enabling fine-grained instance-level editing. It builds a reliable program-supervision generation pipeline and demonstrates improved data efficiency and model performance when pretraining 1B models from scratch.