PreciseEdit: Strong Yet Stable Semantic Transport for One-Step Image Editing
Authors: Jingyi Wang, Xiao Lu
Organizations: Hainan University School of Computer Science and Technology Haikou, China · Shanghai University School of Mechatronic Engineering and Automation Shanghai, China
Recent one-step text-to-image (T2I) models enable efficient image synthesis, opening up new possibilities for interactive Web-based image editing. However, existing one-step editing methods often construct a global transport field through simple vector operations between conditional predictions. While such strong transport can effectively drive the image toward the target semantics, it may be released along unreliable directions or over non-edited regions, leading to structural distortions within the editing region and background drift elsewhere. More importantly, stabilizing the editing process through global smoothing or uniformly weakening the transport field often makes large semantic changes overly conservative. In this work, we revisit one-step image editing from the perspective of the tension between semantic strength and transport stability, and identify a key insight: strong semantic transport is not inherently unstable. What matters is whether it is applied along reliable directions and at reliable spatial locations. Motivated by this insight, we introduce PreciseEdit, a model-agnostic, training-free and inversion-free one-step editing framework that strengthens target-semantic alignment while preserving background fidelity and structural consistency. We further propose a dual-noise-scale role decomposition strategy: a higher-noise scale produces the strong target-directed transport used for actual editing, whereas a lower-noise scale jointly serves as a probe for directional agreement and spatial structure. Extensive experiments on PIE-Bench demonstrate that PreciseEdit achieves state-of-the-art performance and effectiveness on various editing tasks.
Figures & tables
Figure 1. Motivation of PreciseEdit. The first row highlights the importance of reliable spatial localization: unconstrained transport may alter non-target content, while inaccurate localization may miss the intended edit region, raising our first challenge. The second row illustrates the strength–stability dilemma in one-step editing: ChordEdit ( Lu et al., 2026 ) remains conservative even with accurate spatial guidance, whereas the stronger transport of WhereEdit ( Hu et al., 2026 ) achieves larger semantic changes but may affect non-target regions, raising our second challenge. These observations motivate our selective-release principle for strong semantic transport. Motivation examples comparing estimated editing support with ground-truth masks and illustrating PreciseEdit.
Figure 2. Quality–efficiency comparison on PIE-Bench ( Ju et al., 2024 ) . Each point represents an image-editing method. The horizontal axis reports runtime, while the vertical axes report background fidelity measured by PSNR-Unedit (top) and target-semantic alignment measured by CLIP-Edited (bottom) ( Radford et al., 2021 ) . Marker size indicates peak VRAM usage. PreciseEdit achieves a favorable balance among background preservation, semantic alignment, and inference efficiency. Two quality-efficiency scatter plots comparing image editing methods. Runtime is shown on the horizontal axis, PSNR-Unedit and CLIP-Edited are shown on the vertical axes, and marker size represents peak VRAM usage.
Figure 3. Comparison of editing transport properties. (a) Multi-step editing: iterative updates progressively correct the trajectory toward the target. (b) Naive one-step editing: a single conditional residual may provide insufficient target attraction and miss the desired target. (c) ChordEdit ( Lu et al., 2026 ) : temporal smoothing produces a stable low-energy transport field, but can lead to conservative semantic changes. (d) SCT (Ours): anchor-relative target attraction strengthens target-directed transport, while cross-scale directional agreement suppresses unreliable directions for stable one-step editing. Conceptual comparison of multi-step editing, naive one-step editing, ChordEdit, and the proposed SCT with higher- and lower-noise scales.
Figure 4. Overview of PreciseMask. Lower-noise cross-attention produces branch-specific seeds m0,c , which are refined with self-attention affinities to obtain mc⋆ . The source and target supports are combined into S , followed by adaptive extent estimation and soft gating to produce the final PreciseMask M . Pipeline of PreciseMask using cross-attention seeds, self-attention graph construction, graph refinement, and adaptive soft gating.
Type
Method
Consistency
CLIP Semantics
Properties
Efficiency
PSNR ↑
MSE 103↓
LPIPS 103↓
Whole ↑
Edited ↑
T-free
I-free
Step ↓
Runtime(s) ↓
VRAM(MiB) ↓
Multi-step ( ≥30 steps)
DDIM + MasaCtrl ( Song et al., 2021 ; Cao et al., 2023 )
22.19
8.64
105.44
24.02
21.14
✓
×
50
24.49
13016
DDIM + PnP ( Tumanyan et al., 2023 )
18.84
18.42
183.97
24.97
21.87
✓
×
50
16.57
16494
FlowEdit (SD3) ( Kulikov et al., 2025 )
22.18
8.74
105.20
26.81
23.77
✓
✓
33
6.94
18536
Few-step (4 steps)
TurboEdit (SDXL-Turbo) ( Deutch et al., 2024 )
22.50
9.30
105.57
24.68
21.79
✓
✓
4
0.96
18004
InfEdit (SD1.4) ( Xu et al., 2024 )
29.59
2.26
37.10
23.04
20.34
✓
✓
4
0.85
7620
Table 1. Quantitative comparison on PIE-Bench ( Ju et al., 2024 ) . T-free/I-free denote training-/inversion-free. The best/second/third results in each numeric column are highlighted with orange / yellow / blue backgrounds, respectively. The GT-guided oracle row is excluded from ranking.
Figure 5. Comparison of edited results. Real images are in the first column. Prompts are noted under each row. Qualitative examples comparing multi-step, few-step, and one-step image editing methods across several edit types.
Figure 6. Transport-energy localization and editing outcomes. ChordEdit ( Lu et al., 2026 ) yields a comparatively weak field, whereas ungated SCT spreads strong transport beyond the intended region. PreciseMask preserves high transport energy over reliable edit support while suppressing it elsewhere. Transport-energy visualizations and edited outputs comparing ChordEdit, ungated SCT, and PreciseEdit.
Figure 7. Analysis of dual-noise-scale role specialization. Left: Transport-scale analysis. An isolated conditional-residual probe shows that increasing the noise scale strengthens semantic editing but reduces source preservation, indicating greater semantic reach at higher noise. Middle: Localization-scale analysis. Varying only the attention scale for PreciseMask shows that lower-noise attention provides more reliable spatial support; tl=0.35 achieves the best IoU and Boundary F1 while retaining near-maximum AUPRC. Right: Role-allocation analysis. With full PreciseEdit, assigning th=0.90 to semantic transport and tl=0.35 to spatial guidance yields a substantially better fidelity–editability balance than shared-scale or role-swapped alternatives. Three-panel analysis of semantic transport strength, localization quality, and the preservation-editing trade-off under different noise-scale assignments.
Method
PSNR ↑
CLIP-Edited ↑
RunTime(s) ↓
VRAM ↓
IoU(%) ↑
SCT + LIME Mask ( Simsar et al., 2025 )
24.18
21.57
0.74
7450
16.59
SCT + WhereEdit Mask ( Hu et al., 2026 )
23.09
22.89
0.59
7436
44.19
SCT + PreciseMask
24.37
23.36
0.42
6954
53.18
Table 2. Mask-generator comparison under the same SCT field. PreciseMask improves localization and editing quality while reducing runtime and memory relative to LIME ( Simsar et al., 2025 ) and WhereEdit ( Hu et al., 2026 ) .
T2I Backbone
PSNR ↑
CLIP-Edited ↑
ChordEdit
Ours
ChordEdit
Ours
InstaFlow ( Liu et al., 2024 )
21.47
23.95
20.43
21.66
SwiftBrush-v2 ( Dao et al., 2024 )
22.89
23.76
21.23
22.85
SD-Turbo ( Sauer et al., 2024 )
22.64
24.37
22.16
23.36
Table 3. Cross-backbone generalization. PreciseEdit consistently improves both PSNR-Unedit and CLIP-Edited over ChordEdit ( Lu et al., 2026 ) on three fast T2I backbones.
Variant
PSNR ↑
LPIPS ↓
CLIP-Edited ↑
IoU(%) ↑
BF1(%) ↑
A. Major Components (SCT × PreciseMask)
Naive Transport
20.13
–
22.57
–
–
Naive Transport + PreciseMask
27.78
–
21.63
–
–
Ungated SCT
14.56
–
24.25
–
–
Full PreciseEdit
24.37
0.0987
23.36
53.18
42.35
B. SCT (full PreciseMask)
Table 4. Component ablation on PIE-Bench. Block A examines the interaction between the two major components, SCT and PreciseMask. Block B ablates SCT with full PreciseMask fixed, while Block C ablates PreciseMask with full SCT fixed. Best results in each block are bold. “–” denotes metrics not evaluated in the corresponding controlled comparison.
Method
Mask Strategy
PSNR ↑
CLIP-Edited ↑
Naive
w/o Mask
20.13
22.57
w/ GT Mask
31.21
22.23
w/ PreciseMask
27.78
21.63
ChordEdit ( Lu et al., 2026 )
w/o Mask
22.64
22.16
w/ GT Mask
31.38
21.77
w/ PreciseMask
28.55
21.30
Table 5. Effect and transferability of spatial-mask strategies. We insert no mask, the official GT mask, and PreciseMask into the naive, ChordEdit ( Lu et al., 2026 ) , WhereEdit ( Hu et al., 2026 ) , and PreciseEdit transport rules. GT-guided rows provide an oracle upper bound on localization quality.
Recent one-step text-to-image (T2I) models enable efficient image synthesis and provide new opportunities for real-time image editing. However, existing one-step editing methods primarily rely on text conditioning for semantic transformation, lacking explicit spatial control over where to edit. More importantly, even when spatial constraints are introduced, these methods often struggle to achieve strong and stable semantic modifications within the target regions. In this work, we revisit one-step image editing from a spatially controlled perspective and identify two key challenges: discovering editable regions and achieving effective localized semantic transformation. We reveal that existing methods perform global semantic transport, which limits high-intensity local editing under the one-step setting. To address this issue, we propose WhereEdit, a framework that reformulates one-step editing as localized adaptive editing. WhereEdit automatically identifies semantically relevant regions from internal model features and applies adaptive local modulation to enhance target-region editing while preserving non-target areas and structural consistency. Experiments on the PIE-Bench benchmark demonstrate that WhereEdit consistently outperforms existing one-step image editing methods, achieving superior editing quality while maintaining the efficiency of one-step generation. Additional experiments with region-level supervision further highlight the importance of explicit spatial reasoning for high-quality one-step image editing.
Ming Hu, Mingyu Dou, Jianfu Yin +5
Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Zhongnan Hospital of Wuhan University +1
The advent of one-step text-to-image (T2I) models offers unprecedented synthesis speed. However, their application to text-guided image editing remains severely hampered, as forcing existing training-free editors into a single inference step fails. This failure manifests as severe object distortion and a critical loss of consistency in non-edited regions, resulting from the high-energy, erratic trajectories produced by naive vector arithmetic on the models' structured fields. To address this problem, we introduce ChordEdit, a model agnostic, training-free, and inversion-free method that facilitates high-fidelity one-step editing. We recast editing as a transport problem between the source and target distributions defined by the source and target text prompts. Leveraging dynamic optimal transport theory, we derive a principled, low-energy control strategy. This strategy yields a smoothed, variance-reduced editing field that is inherently stable, facilitating the field to be traversed in a single, large integration step. A theoretically grounded and experimentally validated approach allows ChordEdit to deliver fast, lightweight and precise edits, finally achieving true real-time editing on these challenging models.
Liangsi Lu, Xuhang Chen, Minzhe Guo +3
Guangdong University of Technology · Huizhou University · Shenzhen University +1
With recent advancements in large-scale pre-trained text-to-image (T2I) models, training-free image editing methods have demonstrated remarkable success. Typically, these methods involve adding noise to a clean image via an inversion process, followed by separate denoising steps for the reconstruction and editing paths during the forward process. However, since the reconstruction path is approximated using noisy latents from mismatched timesteps, existing methods inevitably suffer from accumulated drift, which fundamentally limits reconstruction fidelity. To address this challenge, we systematically analyze the inversion process within the flow transformer and propose DirectEdit, a simple yet effective editing method that eliminates the inherent reconstruction error without introducing additional neural function evaluations (NFEs). Unlike most prior works that attempt to rectify the inversion path, DirectEdit focuses on directly aligning the forward paths, enabling precise reconstruction and reliable feature sharing. Furthermore, we introduce a preservation mechanism based on attention feature injection and multi-branch mask-guided noise blending, which effectively balances fidelity and editability. Extensive experiments across diverse scenarios demonstrate that DirectEdit achieves efficient and accurate image editing, delivering superior performance that outperforms state-of-the-art methods. Code and examples are available at https://desongyang.github.io/Directedit.
Desong Yang, Mang Ye
National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan, China.