From Scale to Speed: Adaptive Test-Time Scaling for Image Editing

• URL: http://arxiv.org/abs/2603.00141v1
• Date: Tue, 24 Feb 2026 09:13:16 GMT
• Title: From Scale to Speed: Adaptive Test-Time Scaling for Image Editing
• Abstract summary: Image Chain-of-Thought (Image-CoT) is a test-time scaling paradigm that improves image generation by extending inference time.<n>We propose ADaptive Edit-CoT (ADE-CoT), an on-demand test-time scaling framework to enhance editing efficiency and performance.
• Score: 42.704748740631885
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Image Chain-of-Thought (Image-CoT) is a test-time scaling paradigm that improves image generation by extending inference time. Most Image-CoT methods focus on text-to-image (T2I) generation. Unlike T2I generation, image editing is goal-directed: the solution space is constrained by the source image and instruction. This mismatch causes three challenges when applying Image-CoT to editing: inefficient resource allocation with fixed sampling budgets, unreliable early-stage verification using general MLLM scores, and redundant edited results from large-scale sampling. To address this, we propose ADaptive Edit-CoT (ADE-CoT), an on-demand test-time scaling framework to enhance editing efficiency and performance. It incorporates three key strategies: (1) a difficulty-aware resource allocation that assigns dynamic budgets based on estimated edit difficulty; (2) edit-specific verification in early pruning that uses region localization and caption consistency to select promising candidates; and (3) depth-first opportunistic stopping, guided by an instance-specific verifier, that terminates when intent-aligned results are found. Extensive experiments on three SOTA editing models (Step1X-Edit, BAGEL, FLUX.1 Kontext) across three benchmarks show that ADE-CoT achieves superior performance-efficiency trade-offs. With comparable sampling budgets, ADE-CoT obtains better performance with more than 2x speedup over Best-of-N.

Related papers

• Adaptive Inference-Time Scaling via Early-Step Latent Verification for Image Editing [49.40982704974727]
  An unsuitable initial noise may lead to unsatisfactory editing results.<n>Recent inference-time scaling methods address this issue by sampling multiple initial noises and selecting better candidates.<n>We propose VeriLatent, a plug-and-play adaptive inference-time scaling framework with early-step latent verification for image editing.
  arXiv  Detail & Related papers  (2026-06-13T08:30:33Z)
• HABIT: Chrono-Synergia Robust Progressive Learning Framework for Composed Image Retrieval [32.85064790572752]
  Composed Image Retrieval (CIR) is a flexible image retrieval paradigm that enables users to accurately locate the target image.<n>This paper proposes a cHrono-synergiA roBust progressIve learning framework for composed image reTrieval (HABIT)<n>Experiments conducted on two standard CIR datasets demonstrate that HABIT significantly outperforms most methods under various noise ratios.
  arXiv  Detail & Related papers  (2026-04-20T10:02:18Z)
• When, What, and How: Rethinking Retrieval-Enhanced Speculative Decoding [29.402164743559]
  ReSpec is a novel framework that transforms drafter switching into adaptive decision-making.<n>Experiments on Spec-Bench demonstrate that ReSpec state-of-the-art acceleration achieves over $33%$ and $25%$, respectively.
  arXiv  Detail & Related papers  (2025-11-03T06:57:16Z)
• 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)
• ATTS: Asynchronous Test-Time Scaling via Conformal Prediction [112.54016379556073]
  Large language models (LLMs) benefit from test-time scaling but are often hampered by high inference latency.<n>We introduce ATTS (Asynchronous Test-Time Scaling), a statistically guaranteed adaptive scaling framework.<n>We show that ATTS delivers up to 56.7x speedup in test-time scaling and a 4.14x throughput improvement.
  arXiv  Detail & Related papers  (2025-09-18T16:55:09Z)
• From Editor to Dense Geometry Estimator [77.21804448599009]
  We introduce textbfFE2E, a framework that adapts an advanced editing model based on Diffusion Transformer (DiT) architecture for dense geometry prediction.<n>FE2E achieves over 35% performance gains on the ETH3D dataset and outperforms the DepthAnything series, which is trained on 100$times$ data.
  arXiv  Detail & Related papers  (2025-09-04T15:58:50Z)
• In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer [42.44787022143543]
  ICEdit is an in-context editing paradigm with minimal parameter-efficient fine-tuning for quality improvement.<n>It achieves state-of-the-art editing performance with only 0.1% of the training data and 1% trainable parameters compared to previous methods.
  arXiv  Detail & Related papers  (2025-04-29T12:14:47Z)
• Diffusion-Enhanced Test-time Adaptation with Text and Image Augmentation [67.37146712877794]
  IT3A is a novel test-time adaptation method that utilizes a pre-trained generative model for multi-modal augmentation of each test sample from unknown new domains.<n>By combining augmented data from pre-trained vision and language models, we enhance the ability of the model to adapt to unknown new test data.<n>In a zero-shot setting, IT3A outperforms state-of-the-art test-time prompt tuning methods with a 5.50% increase in accuracy.
  arXiv  Detail & Related papers  (2024-12-12T20:01:24Z)
• GraPE: A Generate-Plan-Edit Framework for Compositional T2I Synthesis [10.47359822447001]
  We present an alternate paradigm for T2I synthesis, decomposing the task of complex multi-step generation into three steps.<n>Our approach derives its strength from the fact that it is modular in nature, is training free, and can be applied over any combination of image generation and editing models.
  arXiv  Detail & Related papers  (2024-12-08T22:29:56Z)
• Modality and Task Adaptation for Enhanced Zero-shot Composed Image Retrieval [20.612534837883892]
  Zero-Shot Composed Image Retrieval (ZS-CIR) is designed to retrieve target images using bi-modal (image+text) queries.<n>We propose a lightweight post-hoc framework, consisting of two components.<n>Experiments show that, with the incorporation of our proposed components, inversion-based methods achieve significant improvements.
  arXiv  Detail & Related papers  (2024-10-31T08:49:05Z)
• PostEdit: Posterior Sampling for Efficient Zero-Shot Image Editing [63.38854614997581]
  We introduce PostEdit, a method that incorporates a posterior scheme to govern the diffusion sampling process.<n>The proposed PostEdit achieves state-of-the-art editing performance while accurately preserving unedited regions.<n>The method is both inversion- and training-free, necessitating approximately 1.5 seconds and 18 GB of GPU memory to generate high-quality results.
  arXiv  Detail & Related papers  (2024-10-07T09:04:50Z)

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.