GIDE: 解锁扩散大语言模型用于精确无训练图像编辑
GIDE: Unlocking Diffusion LLMs for Precise Training-Free Image Editing
AI总结：
本文提出GIDE框架，通过离散噪声逆向机制实现无训练图像编辑，支持多种指令并保持背景不变，实验表明其在语义正确性和感知质量上显著优于现有方法。
摘要：
尽管扩散大语言模型（DLLMs）在多模态生成中表现出色，但进行精确无训练图像编辑仍面临挑战。不同于连续扩散模型，DLLMs的离散标记化限制了标准噪声逆向技术的应用，常导致编辑过程中结构退化。本文引入GIDE（Grounded Inversion for DLLM Image Editing），一种统一框架，旨在弥合这一差距。GIDE包含一种新的离散噪声逆向机制，能够准确捕捉离散标记空间中的潜在噪声模式，确保高保真重建。随后，我们将编辑流程分解为接地、逆向和细化阶段。此设计使GIDE支持多种编辑指令（文本、点和框）和操作，同时严格保持未编辑的背景。此外，为克服现有单步评估协议的限制，我们引入GIDE-Bench，一个包含805个由多样化多模态输入引导的组合编辑场景的严格基准。在GIDE-Bench上的广泛实验表明，GIDE在语义正确性上比现有无训练方法提高了51.83%，在感知质量上提高了50.39%。此外，在ImgEdit-Bench上的额外评估证实了其广泛的应用性，证明了其在训练基线上的持续优势，并在与领先模型相当的逼真一致性上取得了成果。
英文摘要：
While Diffusion Large Language Models (DLLMs) have demonstrated remarkable capabilities in multi-modal generation, performing precise, training-free image editing remains an open challenge. Unlike continuous diffusion models, the discrete tokenization inherent in DLLMs hinders the application of standard noise inversion techniques, often leading to structural degradation during editing. In this paper, we introduce GIDE (Grounded Inversion for DLLM Image Editing), a unified framework designed to bridge this gap. GIDE incorporates a novel Discrete Noise Inversion mechanism that accurately captures latent noise patterns within the discrete token space, ensuring high-fidelity reconstruction. We then decompose the editing pipeline into grounding, inversion, and refinement stages. This design enables GIDE supporting various editing instructions (text, point and box) and operations while strictly preserving the unedited background. Furthermore, to overcome the limitations of existing single-step evaluation protocols, we introduce GIDE-Bench, a rigorous benchmark comprising 805 compositional editing scenarios guided by diverse multi-modal inputs. Extensive experiments on GIDE-Bench demonstrate that GIDE significantly outperforms prior training-free methods, improving Semantic Correctness by 51.83% and Perceptual Quality by 50.39%. Additional evaluations on ImgEdit-Bench confirm its broad applicability, demonstrating consistent gains over trained baselines and yielding photorealistic consistency on par with leading models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
