RegionCache: Semantic-Aware Region Reuse for Efficient Multi-Turn Image Generation

• URL: http://arxiv.org/abs/2608.29809v1
• Date: Sun, 30 Aug 2026 14:21:46 GMT
• Title: RegionCache: Semantic-Aware Region Reuse for Efficient Multi-Turn Image Generation
• Abstract summary: RegionCache is a semantic-aware reuse framework for multi-turn image editing.<n>It selectively reuses diffusion states from unchanged regions.<n>It achieves 1.43x--2.55x end-to-end speedup while maintaining comparable image quality.
• Score: 10.261275777311633
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Real-world image generation often involves multi-turn editing, where users iteratively modify small regions while most image content remains unchanged. However, existing diffusion transformer (DiT)-based editing pipelines recompute the entire image at every turn, causing substantial redundant computation. Existing DiT acceleration methods further ignore semantic correspondence across prompts, leading to unnecessary recomputation or unsafe reuse that harms editing quality. To address this, we propose RegionCache, a semantic-aware reuse framework for multi-turn image editing that selectively reuses diffusion states from unchanged regions. RegionCache detects reusable regions through semantic overlap between consecutive prompts and cross-attention localization, and adopts an adaptive reuse schedule based on prompt similarity and contextual consistency. Experiments on PixArt-alpha demonstrate that RegionCache achieves 1.43x--2.55x end-to-end speedup while maintaining comparable image quality. Code is available at https://github.com/hebutBryant/RegionCache.

Related papers

• WorldCache: Content-Aware Caching for Accelerated Video World Models [50.7543797435026]
  We present textbfWorldCache, a Perception-Constrained Dynamic CaCache framework.<n>WorldCache improves both when and how to reuse features.<n>On Cosmos-2.5-2B evaluated on PAI-Bench, WorldCache achieves textbf$2.3times$ inference speedup while preserving textbf99.4% of baseline quality.
  arXiv  Detail & Related papers  (2026-03-23T17:59:54Z)
• SpotEdit: Selective Region Editing in Diffusion Transformers [66.44912649206553]
  SpotEdit is a training-free diffusion editing framework that selectively updates only the modified regions.<n>By reducing unnecessary computation and maintaining high fidelity in unmodified areas, SpotEdit achieves efficient and precise image editing.
  arXiv  Detail & Related papers  (2025-12-26T14:59:41Z)
• RegionE: Adaptive Region-Aware Generation for Efficient Image Editing [28.945176886517448]
  RegionE is an adaptive, region-aware generation framework that accelerates IIE tasks without additional training.<n>The framework consists of three main components: 1) Adaptive Region Partition, 2) Region-Aware Generation, and 3) Adaptive Velocity Decay Cache.<n>We applied RegionE to state-of-the-art IIE base models, including Step1X-Edit, FLUX.1 Kontext, and Qwen-Image-Edit.
  arXiv  Detail & Related papers  (2025-10-29T14:58:37Z)
• Training-free Geometric Image Editing on Diffusion Models [53.38549950608886]
  We tackle the task of geometric image editing, where an object within an image is repositioned, reoriented, or reshaped.<n>We propose a decoupled pipeline that separates object transformation, source region inpainting, and target region refinement.<n>Both inpainting and refinement are implemented using a training-free diffusion approach, FreeFine.
  arXiv  Detail & Related papers  (2025-07-31T07:36:00Z)
• EEdit: Rethinking the Spatial and Temporal Redundancy for Efficient Image Editing [47.68813248789496]
  We propose a framework, named EEdit, to achieve efficient image editing.<n>Experiments demonstrate an average of 2.46 $times$ acceleration without performance drop in a wide range of editing tasks.
  arXiv  Detail & Related papers  (2025-03-13T11:26:45Z)
• SmoothCache: A Universal Inference Acceleration Technique for Diffusion Transformers [4.7170474122879575]
  Diffusion Transformers (DiT) have emerged as powerful generative models for various tasks, including image, video, and speech synthesis.<n>We introduce SmoothCache, a model-agnostic inference acceleration technique for DiT architectures.<n>Our experiments demonstrate that SmoothCache achieves 71% 8% to speed up while maintaining or even improving generation quality across diverse modalities.
  arXiv  Detail & Related papers  (2024-11-15T16:24:02Z)
• FastDrag: Manipulate Anything in One Step [20.494157877241665]
  We introduce a novel one-step drag-based image editing method, i.e., FastDrag, to accelerate the editing process. This innovation achieves one-step latent semantic optimization and hence significantly promotes editing speeds. Our FastDrag is validated on the DragBench dataset, demonstrating substantial improvements in processing time over existing methods.
  arXiv  Detail & Related papers  (2024-05-24T17:59:26Z)
• Accelerating Text-to-Image Editing via Cache-Enabled Sparse Diffusion Inference [36.73121523987844]
  We introduce Fast Image Semantically Edit (FISEdit), a cached-enabled sparse diffusion model inference engine for efficient text-to-image editing. FISEdit uses semantic mapping between the minor modifications on the input text and the affected regions on the output image. For each text editing step, FISEdit can automatically identify the affected image regions and utilize the cached unchanged regions' feature map to accelerate the inference process.
  arXiv  Detail & Related papers  (2023-05-27T09:14:03Z)
• Region-Aware Diffusion for Zero-shot Text-driven Image Editing [78.58917623854079]
  We propose a novel region-aware diffusion model (RDM) for entity-level image editing. To strike a balance between image fidelity and inference speed, we design the intensive diffusion pipeline. The results show that RDM outperforms the previous approaches in terms of visual quality, overall harmonization, non-editing region content preservation, and text-image semantic consistency.
  arXiv  Detail & Related papers  (2023-02-23T06:20:29Z)
• Zero-shot Image-to-Image Translation [57.46189236379433]
  We propose pix2pix-zero, an image-to-image translation method that can preserve the original image without manual prompting. We propose cross-attention guidance, which aims to retain the cross-attention maps of the input image throughout the diffusion process. Our method does not need additional training for these edits and can directly use the existing text-to-image diffusion model.
  arXiv  Detail & Related papers  (2023-02-06T18:59:51Z)

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.