UniSlider: Perceptually Uniform Sliders for Continuous Image Editing

David Serrano-Lozano; Duygu Ceylan; Yannick Hold-Geoffroy; Iliyan Georgiev; Javier Vazquez-Corral; Anna Frühstück

cs.CVOct 5, 2026

UniSlider: Perceptually Uniform Sliders for Continuous Image Editing

Authors: David Serrano-Lozano, Duygu Ceylan, Yannick Hold-Geoffroy, Iliyan Georgiev, Javier Vazquez-Corral, Anna Frühstück

Organizations: Computer Vision Center · Universitat Autònoma de Barcelona · Adobe Research

Abstract

Sliders provide an intuitive interface for continuous image editing. In current generative approaches, however, the slider is simply a rescaling of the method's strength parameter, such as an adapter coefficient, a prompt weight, or an interpolation factor. This strength relates poorly to perceptual change. The image can partially revert as the slider moves, long stretches of the range produce no visible difference, and short intervals transform the image abruptly. Remapping the strength could fix this uneven pace, but only if the trajectory is monotone, which current methods do not enforce. We therefore distinguish the slider from the strength, and require perceptual distance from the input to grow linearly with the slider value. We introduce UniSlider, a lightweight LoRA trained on a few-step editing backbone so that its strength approximates this ideal slider. Few-step sampling lets us impose this objective in pixel space without intermediate ground truth, and the backbone's output is preserved at full strength. However, a low-rank adapter cannot make the strength fully uniform. Our slider is thus an inference-time remapping of the strength, obtained by adaptive sampling. Since training optmizes to make the trajectory monotone, this remapping closes the remaining gap without extra training or parameters. On a new benchmark of 300 continuous edits evaluating uniformity, monotonicity, edit fidelity, and identity preservation, UniSlider outperforms all prior methods and is preferred in a user study.

Figures & tables

Panel of Figure 1 — Figure 1: UniSlider produces perceptually uniform edit strengths. Each row shows one image edited at increasing strength under the instruction above it. Equidistant steps along the slider produce equal amounts of perceptual change; the full interval is thus utilized smoothly and predictably.

Figure 2: Training closes most of the gap to a uniform slider; adaptive sampling closes the rest. All three rows edit the same image with the same instruction. Left: frames at equal slider values for SliderEdit, our loss alone ( Vanilla ), and our full UniSlider method with strength reparametrization. Right: perceptual distance from the input to every slider frame, normalized by the total input–edit distance. The diagonal marks a perfectly uniform slider. We use DreamSim ( Fu et al., 2023 ) as a perceptual metric.

Figure 3: UniSlider training scheme. We sample $s$ , scale the trainable LoRA against the frozen backbone ( Eq. 3 ), and run the full $T$ -step inference process to get $x_{s}$ ; $x_{\mathrm{edit}}$ is cached once at $s{=}1$ . The loss ( Eq. 4 ) pushes $x_{s}$ to the $s$ -implied perceptual position along the path from $x_{\mathrm{src}}$ to $x_{\mathrm{edit}}$ .

Panel of Figure 4 — Figure 4: Uniformity degrades with the number of training examples, and is recovered by adaptive sampling. Normalized DreamSim distance from the input to the output, for adapter rank $r$ and $N$ overfitting examples. The diagonal is the ideal uniform slider. First six columns: the strength is the slider ( $s=u$ ). Rightmost column (full method): one adapter trained on the full dataset, evaluated on $N=32$ unseen examples, with adaptive sampling mapping $u$ to $u=g(s)$ .

	Edit Fidelity		Identity	Endpoint	IQA	Monotonicity		Uniformity	User Study
Method	CLIP dir $\uparrow$	VQA edit $\downarrow$	VQA id $\downarrow$	RMSE $\downarrow$	MUSIQ $\uparrow$	$\mathcal{V}_{\text{VQA}}\!\downarrow$	$\mathcal{V}_{\text{LPIPS}}\!\downarrow$	CV-DSim $\downarrow$	$\bigstar\uparrow$	$\equiv\uparrow$
K. Kontext	0.078	0.430	0.161	0.021	65.34	0.047	0.116	1.153	40.3%	46.8%
Diff. Steering	0.107	0.710	0.167	0.151	64.59	0.094	0.390	1.045	61.4%	43.9%
FlowSlider	0.090	0.322	0.378	0.047	65.84	0.067	0.165	0.675	14.1%	27.0%
VeloEdit	0.079	0.470	0.151	0.023	53.18	0.072	0.078	0.996	57.7%	61.7%
SliderEdit	0.087	0.440	0.103	0.048	63.70	0.047	0.251	1.199	42.2%	35.6%

Table 1: Comparison against baselines. All metrics are averaged over the 300 benchmark examples. Shading marks the best and second-best method per column.

\bigstar

marks the method preferred overall in the user study and

\equiv

the one preferred for uniformity.

Panel of Figure 5 — Figure 5: UniSlider results. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B.

Panel of Figure 6 — Figure 6: Composing a masked edit with a global one. A $4{\times}4$ grid over two instructions applied to the same input. Strength increases left to right and top to bottom. The second axis is obtained by re-running UniSlider on the output of the first, no composed or masked examples appear in training.

Table 8

Appendix figures & tables16 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 7: Backbone and DreamSim failure case. The backbone’s default edit changes the colors of the ball, and this change is carried along the whole slider. DreamSim places the second frame on the diagonal because of a small color change on the right of the ball, although the ball is not yet inflated.

Figure 8: Adaptive sampling vs. dense reference. Left: five of the eleven frames obtained with adaptive sampling (top) and with the dense reference, which inverts a PCHIP spline fitted to 51 measurements (bottom). Right: normalized DreamSim distance from the input to each frame against the slider value $u$ . Without remapping (dashed), adaptive sampling (blue) and dense fit (red).

Panel of Figure 9 — Figure 9: Training data. Nine of the $300$ (image, instruction) pairs. These two things are the entire supervision, the target at $s{=}1$ is the frozen backbone’s default edit, so no edited image and no intermediate is required. Instructions carry no degree or mechanism (no “slightly”, no “by 30%”), since strength is carried entirely by $s$ ; they are correspondingly short, a median of eight words.

		Edit Fidelity		Identity	Endpoint	IQA	Monotonicity		Uniformity
	Method	CLIP dir $\uparrow$	VQA edit $\downarrow$	VQA id $\downarrow$	RMSE $\downarrow$	MUSIQ $\uparrow$	$\mathcal{V}_{\text{VQA}}\!\downarrow$	$\mathcal{V}_{\text{LPIPS}}\!\downarrow$	CV-DSim $\downarrow$
Global	K. Kontext	0.097	0.414	0.076	0.020	62.86	0.042	0.061	0.985
	Diff. Steering	0.124	0.660	0.025	0.182	61.91	0.128	0.348	0.986
	FlowSlider	0.095	0.242	0.143	0.047	64.44	0.045	0.167	0.673
	VeloEdit	0.106	0.470	0.053	0.021	54.19	0.070	0.022	0.782
	SliderEdit	0.106	0.412	0.037	0.081	62.78	0.068	0.132	1.000

Appendix

Table 4: Benchmark results by edit type. The same metrics as Tab. 1 , split over the three edit categories (

100

examples each,

300

total). K marks a method running on the FLUX.2-klein-4B backbone, i.e. the same base as ours, separating backbone advantage from method.

	Edit Fid.	IQA	Monotonicity		Uniformity	Fidelity $\times$ Preservation
Method	QEval	CLIPIQA $\uparrow$	$\mathcal{V}_{\text{DreamSim}}\!\downarrow$	$\mathcal{V}_{\text{Qeval}}\!\downarrow$	CV-TPIPS $\downarrow$	VQA ${}_{\mathrm{edit}}\times(1-$ VQA ${}_{\mathrm{id}})$ $\uparrow$
KontinuousKontext	0.660	0.488	0.0007	0.167	0.438	0.357
DiffusionSteering	0.759	0.489	0.0008	0.196	0.468	0.592
FlowSlider	0.607	0.503	0.0004	0.175	0.420	0.196
VeloEdit	0.682	0.402	0.0005	0.168	0.427	0.407
SliderEdit	0.671	0.469	0.0009	0.147	0.433	0.386

Appendix

Table 5: Additional metrics. Averaged over the 300 benchmark examples. Averaged over all benchmark examples.

Panel of Figure 10 — Figure 10: Benchmark 2AFC user study. Left: a screenshot of the study interface, in which annotators drag a slider back and forth to compare methods A and B at matching strengths. Right: Thurstone Case V values with confidence intervales from the pairwise judgments, for the two questions asked: which method is better overall, and which is more uniform. A total of 25 observers participated in the study making a total of 881 examples.

Figure 11: Perceptual metric 2AFC user study. Left: two examples of the rendered images at different strengths. Right: PLCC and SRCC for all evaluated metrics. A total of 15 observers participated in the study making a total of 538 annotations.

Panel of Figure 12 — Figure 12: Benchmark comparisons with per-frame metrics. The first two examples compare UniSlider against Firefly Tune (top) and against FlowSlider and DiffusionSteering (second), annotating each frame with its $\mathrm{VQA}_{\text{edit}}$ and $\mathrm{VQA}_{\text{id}}$ values. The last two compare UniSlider against Firefly Tune and SliderEdit, plotting instead the perceptual distance from the input at each frame, so that a flat segment marks a dead zone and a steep one a jump. Note how UniSlider achieves better uniformity across all different types of edits.

Panel of Figure 13 — Figure 13: Additional UniSlider results with changes in visual style. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 14 — Figure 14: Additional UniSlider results with attribute edits. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 15 — Figure 15: Additional UniSlider results with structural edits. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 16 — Figure 16: Additional UniSlider results with structural edits. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 17 — Figure 17: Additional UniSlider with masking. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 18 — Figure 18: Additional 2D UniSlider results. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

Panel of Figure 19 — Figure 19: Additional 2D UniSlider results. From left to right input image, all the intermediate strength levels and the default output of Flux-Klein-4B. All the input images are from the DIV2K ( Agustsson & Timofte, 2017 ) dataset.

	Edit Fidelity		Identity	Endpoint	IQA	Monotonicity		Uniformity	Efficiency
	CLIP dir $\uparrow$	VQA edit $\downarrow$	VQA id $\downarrow$	RMSE $\downarrow$	MUSIQ $\uparrow$	$\mathcal{V}_{\text{VQA}}\!\downarrow$	$\mathcal{V}_{\text{LPIPS}}\!\downarrow$	CV-DSim $\downarrow$	Time(s)	Param
Qwen-Edit-Dist	0.102	0.693	0.158	0.059	67.21	0.055	0.031	0.641	32.4	29B
FireRed-1.0-Dist	0.109	0.802	0.114	0.183	67.93	0.089	0.052	1.038	68.7	29B
Flux-Klein-4B	0.112	0.799	0.125	0.018	67.09	0.046	0.016	0.652	7.2	8B
Flux-Klein-9B	0.114	0.820	0.141	0.018	68.19	0.051	0.018	0.748	14.8	17B
$r=4$	0.112	0.799	0.125	0.020	66.94	0.052	0.018	0.661	7.2	1M

Appendix

Table 6: Editing backbone and LoRA rank ablations. Top: UniSlider trained on four editing backbones. Bottom: adapter rank swept on Flux-Klein-4B, our default setting (

r{=}32

, shown in both blocks). Time is the wall-clock cost of generating an

11

-frame slider on an RTX Pro 6000; Params counts the full model in the top block and the trainable adapter in the bottom one.

Explore similar work

Scalable Minimal-Change Learning for Controllable Image Editing
Oct 5, 2026Shuo Chen, Fengming Huang, Yu Yao +2Multi-Reference Image Generation
Sydney AI Centre, The University of Sydney
GeoEdit: Local Frames for Fast, Training-Free On-Manifold Editing in Diffusion Models
Apr 27, 2026Yiming Zhang, Sitong Liu, Ke Li +3Diffusion-Based Image Editing Diffusion Models
University of California San Diego, La Jolla, CA, USA · University of Washington, Seattle, WA, USA · Xidian University, Xi’an, China
ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
Apr 22, 2026Shelly Golan, Michael Finkelson, Ariel Bereslavsky +2Multi-Objective Reinforcement Learning Diffusion Models
Tel Aviv University · Lightricks · Adobe Research

Oct 5, 2026, cs.CVJ/K move · Enter open · S save

Scalable Minimal-Change Learning for Controllable Image Editing

Shuo Chen, Fengming Huang, Yu Yao, Mingming Gong+1

Sydney AI Centre, The University of Sydney

Multi-Reference Image Generation

Open paper AskPDF arXiv

Image editing should change only the attributes specified by an instruction while preserving everything else, yet current methods often make unintended changes. We treat this minimal-change principle as an optimization objective for instruction-based editing. Latent L1 regularization is a poor proxy for output locality in modern nonlinear generators and often requires supervision unavailable at scale. We instead optimize edit outcomes with reinforcement learning. An agentic vision-language reward model audits each source image, instruction, and edited image for two failure types: unimplemented requested changes and unintended changes. A group-level rubric merges and verifies these issues to provide consistent rewards across candidate edits without per-instruction human annotations. On FLUX.1 Kontext-dev, ARRO raises average EditScore from 5.21 to 5.88 across MinEval, MagicBrush, AnyBench, and Emu-Edit. On 600 evaluation examples, it reduces off-target pixel change by 8.4% relative to the base editor. Reward and SFT controls, blinded human evaluations, and transfer to OmniGen2 provide complementary evidence. Code: https://github.com/Showwwwwwwww/ARRO

Abstract

Explore similar work

Scalable Minimal-Change Learning for Controllable Image Editing

GeoEdit: Local Frames for Fast, Training-Free On-Manifold Editing in Diffusion Models

ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control