On the Controllability-Fidelity Frontier in Diffusion Editing

• URL: http://arxiv.org/abs/2606.09901v1
• Date: Fri, 05 Jun 2026 13:24:01 GMT
• Title: On the Controllability-Fidelity Frontier in Diffusion Editing
• Abstract summary: We study the trade-offs between adherence to user intent, preservation of non-target content, and output quality.<n>Our work spans text- and mask-guided edits, point/drag manipulation, and inversion-based pipelines.<n>We propose algorithmic frameworks (with pseudocode) for mask-localized and instruction-guided editing.
• Score: 7.759950850415493
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion-based generative models enable powerful image editing capabilities, but achieving precise control while maintaining fidelity and safety remains challenging. We present a comprehensive theoretical and empirical study of controllable diffusion-based image editing, analyzing the trade-offs between adherence to user intent, preservation of non-target content, and output quality. Our work spans text- and mask-guided edits, point/drag manipulation, and inversion-based pipelines. We derive mathematical formulations of editing objectives and analyze dynamics of noise injection, score guidance, and inversion error. We provide theoretical bounds on reconstruction error, stability under repeated edits, and locality of changes. We propose algorithmic frameworks (with pseudocode) for mask-localized and instruction-guided editing, and present extensive experiments comparing state-of-the-art methods (e.g.\ TF-ICON \cite{lu2023tficone}, DragFlow \cite{zhou2025dragflow}, InstructPix2Pix \cite{brooks2023instructpix2pix}, UltraEdit \cite{zhao2024ultraedit}) on multiple tasks and metrics (FID, identity similarity, CLIP alignment, artifact scores, etc). Our results reveal key failure modes, such as identity drift, prompt sensitivity, and compositional errors. We also discuss ethical considerations in image editing, including misuse risks, bias, consent, and concept erasure techniques (e.g.\ MACE \cite{lu2024mace}, ANT \cite{li2025ant}, EraseAnything \cite{gao2024eraseanything}) as safeguards. We conclude with best practices and future directions for responsible, high-fidelity diffusion-based editing.

Related papers

• Editing Away the Evidence: Diffusion-Based Image Manipulation and the Failure Modes of Robust Watermarking [0.3644540605358075]
  diffusion-based image editing injects noise and reconstructs images through a powerful generative prior.<n>We show that non-adversarial diffusion editing can unintentionally degrade or remove robust watermarks.<n>Results indicate that even routine semantic edits can significantly reduce watermark recoverability.
  arXiv  Detail & Related papers  (2026-03-13T12:46:27Z)
• The Devil is in Attention Sharing: Improving Complex Non-rigid Image Editing Faithfulness via Attention Synergy [71.39358554558667]
  We introduce SynPS, a method that Synergistically leverages Positional embeddings and Semantic information for faithful non-rigid image editing.<n>We propose an editing measurement that quantifies the required editing magnitude at each denoising step.<n>By adaptively integrating positional and semantic cues, SynPS effectively avoids both over- and under-editing.
  arXiv  Detail & Related papers  (2025-12-16T14:08:00Z)
• EditInfinity: Image Editing with Binary-Quantized Generative Models [64.05135380710749]
  We investigate the parameter-efficient adaptation of binary-quantized generative models for image editing.<n>Specifically, we propose EditInfinity, which adapts emphInfinity, a binary-quantized generative model, for image editing.<n>We propose an efficient yet effective image inversion mechanism that integrates text prompting rectification and image style preservation.
  arXiv  Detail & Related papers  (2025-10-23T05:06:24Z)
• Immunizing Images from Text to Image Editing via Adversarial Cross-Attention [17.498230426195114]
  We propose a novel attack that targets the visual component of editing methods.<n>We introduce Attention Attack, which disrupts the cross-attention between a textual prompt and the visual representation of the image.<n> Experiments conducted on the TEDBench++ benchmark demonstrate that our attack significantly degrades editing performance while remaining imperceptible.
  arXiv  Detail & Related papers  (2025-09-12T15:47:50Z)
• Flux-Sculptor: Text-Driven Rich-Attribute Portrait Editing through Decomposed Spatial Flow Control [24.057959328107387]
  Flux-Sculptor is a flux-based framework designed for precise text-driven portrait editing.<n>Our framework introduces a Prompt-Aligned Spatial Locator (PASL) to accurately identify relevant editing regions.<n>Experiments demonstrate that Flux-Sculptor surpasses existing methods in rich-attribute editing and facial information preservation.
  arXiv  Detail & Related papers  (2025-07-05T10:08:39Z)
• Concept Lancet: Image Editing with Compositional Representation Transplant [58.9421919837084]
  Concept Lancet is a zero-shot plug-and-play framework for principled representation manipulation in image editing.<n>We decompose the source input in the latent (text embedding or diffusion score) space as a sparse linear combination of the representations of the collected visual concepts.<n>We perform a customized concept transplant process to impose the corresponding editing direction.
  arXiv  Detail & Related papers  (2025-04-03T17:59:58Z)
• Training-Free Text-Guided Image Editing with Visual Autoregressive Model [46.201510044410995]
  We propose a novel text-guided image editing framework based on Visual AutoRegressive modeling.<n>Our method eliminates the need for explicit inversion while ensuring precise and controlled modifications.<n>Our framework operates in a training-free manner and achieves high-fidelity editing with faster inference speeds.
  arXiv  Detail & Related papers  (2025-03-31T09:46:56Z)
• Lost in Edits? A $λ$-Compass for AIGC Provenance [119.95562081325552]
  We propose a novel latent-space attribution method that robustly identifies and differentiates authentic outputs from manipulated ones.<n>LambdaTracer is effective across diverse iterative editing processes, whether automated through text-guided editing tools such as InstructPix2Pix or performed manually with editing software such as Adobe Photoshop.
  arXiv  Detail & Related papers  (2025-02-05T06:24:25Z)
• Doubly Abductive Counterfactual Inference for Text-based Image Editing [130.46583155383735]
  We study text-based image editing (TBIE) of a single image by counterfactual inference. We propose a Doubly Abductive Counterfactual inference framework (DAC) Our DAC achieves a good trade-off between editability and fidelity.
  arXiv  Detail & Related papers  (2024-03-05T13:59:21Z)
• DiffEditor: Boosting Accuracy and Flexibility on Diffusion-based Image Editing [66.43179841884098]
  Large-scale Text-to-Image (T2I) diffusion models have revolutionized image generation over the last few years. We propose DiffEditor to rectify two weaknesses in existing diffusion-based image editing. Our method can efficiently achieve state-of-the-art performance on various fine-grained image editing tasks.
  arXiv  Detail & Related papers  (2024-02-04T18:50:29Z)
• Unified Diffusion-Based Rigid and Non-Rigid Editing with Text and Image Guidance [15.130419159003816]
  We present a versatile image editing framework capable of executing both rigid and non-rigid edits. We leverage a dual-path injection scheme to handle diverse editing scenarios. We introduce an integrated self-attention mechanism for fusion of appearance and structural information.
  arXiv  Detail & Related papers  (2024-01-04T08:21:30Z)
• iEdit: Localised Text-guided Image Editing with Weak Supervision [53.082196061014734]
  We propose a novel learning method for text-guided image editing. It generates images conditioned on a source image and a textual edit prompt. It shows favourable results against its counterparts in terms of image fidelity, CLIP alignment score and qualitatively for editing both generated and real images.
  arXiv  Detail & Related papers  (2023-05-10T07:39:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.