Efficient and Faithful High-Resolution Image Editing with Sparse Cross-Resolution Denoising
cs.CV
Submitted: 2026-11-03
Updated: 2026-11-03
Code: https://github.com/FDUPeter/FLASH
License: http://creativecommons.org/licenses/by/4.0/
The gist: Instruction-based image editors still primarily operate at 1K, whereas practical editing increasingly requires faithful 2K/4K outputs.
Terminology
Abstract
Instruction-based image editors still primarily operate at 1K, whereas practical editing increasingly requires faithful 2K/4K outputs. Dense high-resolution adaptation preserves the editing trajectory but incurs expensive target-source attention; editing at 1K followed by external refinement is cheaper but can separate detail restoration from the editing trajectory. We introduce FLASH, a native high-resolution editor based on sparse cross-resolution denoising. A frozen 1K branch provides a semantic and structural anchor, while a target-resolution branch reconstructs local detail within the same Transformer. An asymmetric topology retains full 1K attention, guides high-resolution queries with spatially aligned low-resolution noisy-target states, and confines high-resolution target-reference interaction to local Halo neighborhoods. Resolution-specific diffusion-state mapping and spatial YaRN adapt the pretrained editor to larger grids. An optional, training-free progressive path replaces early high-resolution denoising with a 1K prefix and completes refinement in the same editor. On VINS-4KEval, the direct model achieves the strongest structural fidelity among the compared methods at both 2K and 4K. A separate 4K forward benchmark with R = 1 records 4.2x faster total forward execution and 25.3x faster attention than dense target-reference attention. Progressive inference reduces end-to-end 4K latency by 30.5%, with a modest quality cost.
Sources
- RegionE: Adaptive Region-Aware Generation for Efficient Image Editing
- On the Importance of Noise Scheduling for Diffusion Models
- VINS-120K: Ultra High-Resolution Image Editing with A Large-Scale Dataset
- Scalable High-Resolution Pixel-Space Image Synthesis with Hourglass Diffusion Transformers
- LoRA: Low-Rank Adaptation of Large Language Models
- FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
- Flow Matching for Generative Modeling
- Step1X-Edit: A Practical Framework for General Image Editing
- Qwen-Image Technical Report
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs