Papers for

graphic design software developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

ProtoDiffusion speeds up layered graphic design generation and editing

ProtoDiffusion: Token-per-Layer Generative Modeling of Multi-Layer Images

Abstract: Graphic designs are ordered compositions of RGBA layers whose separation enables selection, movement, and reuse. Existing layered models represent each layer with a full-canvas or region-sized spatial latent grid, so representation size grows with the cumulative spatial extent of the layers. We present ProtoDiffusion, a latent-diffusion framework built on ProtoVAE, a multi-layer variational autoencoder that encodes a design into one shared spatial prototype and one candidate vector per layer. Each candidate conditions a shared decoder to reconstruct its corresponding RGBA layer. For $N$ layers, the representation contains $HW+N$ tokens before backbone-specific packing, compared with $NHW$ for full-grid-per-layer representations. ProtoDiffusion jointly denoises the prototype and a fixed set of candidate slots, conditioned on an image or text, with bounding boxes as optional layout guidance. The model leaves unneeded slots empty and determines the number of layers and, in absence of boxes, their layout, requiring neither as input. The same architecture supports both image-to-design and text-to-design. ProtoVAE achieves high-fidelity reconstruction from its compact shared latent representation, reaching 40.43 dB PSNR on layer crops. On the LayerD evaluation split, image-to-design achieves RGB L1 scores of 0.040 with bounding boxes and 0.057 without, with box-guided alpha soft IoU (0.850) comparable to the strongest fine-tuned baseline (0.871). On merged images, text-to-design reaches 66.41 FID versus 86.84 for FLUX.2 [klein], achieves a 0.860 text-accuracy score, and runs $19.1\times$ faster than the same-backbone generation-then-decomposition cascade.

Thu 19 NovArtificial Intelligence
The gist
Creating graphics often involves stacking layers of images and colors, which can get large and complex. The authors developed ProtoDiffusion, a new way to represent and generate these layers more efficiently by sharing some parts of the image and only separately handling the parts that change. This approach makes it faster to create, edit, and rearrange layers from images or text descriptions. It also figures out how many layers are needed and where they go without extra instructions, making design generation more flexible.
Open → 2611.26586v1

Controlling feature injection improves image editing trade offs

Inject Where It Matters: Objective-Dependent Feature Control in Rectified-Flow Image Editing

Abstract: Many prompt-driven FLUX editing methods rely on injecting source features into target-conditioned generation, creating an inherent trade-off between source preservation and the desired edit. We address this problem by controlling where source features are injected and which transformer blocks are used. First, we introduce adaptive masks that spatially gate feature injection using differences between source- and target-conditioned block outputs at matched inversion states. Second, we use REINFORCE to select Top-K injection blocks under different preservation-editability objectives. The selected blocks are fixed at inference and require no online search. On PIE-Bench with FLUX.1-dev, spatial masking controls the alignment-preservation trade-off. On 60 validation examples, a learned nine-block set gives lower whole-image LPIPS than the 18-block RF-Edit set within the same masked pipeline, with slightly lower alignment. Among 50 random nine-block sets, it has the lowest LPIPS but not the highest reward. Separate objective sweeps reveal different block preferences for preservation and alignment.

Thu 19 NovArtificial Intelligence
The gist
Image editing tools often struggle to balance keeping the original image details while making desired changes. The authors introduce a way to carefully choose where and how features from the original image are applied during editing. They use adaptive masks and a selection method to pick parts of the processing pipeline that best match editing goals. This approach helps keep the image looking right while still making effective edits. The results show better overall image quality and flexible control over preserving or changing details.
Open → 2611.25811v1

SketchBloom expands sparse sketches into detailed editable vector art

SketchBloom: Training-Free Vector Sketch Outpainting with Stroke-Level Style Control

Abstract: Outpainting sparse sketches while preserving contour coherence and stroke style remains challenging. We present SketchBloom, a training-free framework that combines pretrained raster outpainting and style-transfer guidance with per-instance vector optimization. Given a sketch, a scene prompt, and an optional reference style, it produces an editable Scalable Vector Graphic (SVG) composed of Bezier strokes. Adaptive latent blending balances content contours and reference texture without model retraining. A reference-derived stroke vocabulary initializes contour- and DINO-attention-guided strokes, which are optimized separately for structure and texture before contour-emphasizing multiply fusion. As in prior optimization-based vector sketching, no task-specific network is trained; our contribution is the synthesis design specialized to sketch outpainting. On LineArtOut-1K, comprising 1,024 input and style combinations, SketchBloom improves perceptual, semantic, and structural fidelity over the evaluated vector baselines when all methods vectorize the same outpainted raster. This protocol evaluates vector synthesis rather than end-to-end outpainting or reference-style superiority. Runtime is approximately six minutes per pair on a single NVIDIA H200 GPU. Project page: https://khangcntn123.github.io/SketchBloom/

Fri 30 OctComputer Vision and Pattern Recognition
The gist
Drawing more of a simple sketch while keeping the style and lines consistent is hard. The authors created SketchBloom, a tool that extends sketches without needing new training. It uses existing image outpainting and style transfer methods combined with special stroke controls to create detailed, editable vector drawings. This means you can add new parts to your sketch that match the original style and can still be adjusted easily.
Open → 2611.00245v1

Video stylization method adds style to chosen regions with temporal consistency

LociArt: A Unified Paradigm for Free-Style Local Video Stylization

Abstract: Video stylization typically transforms entire scenes, while local artistic editing requires precise region--style association, spatial control, and temporal consistency. We introduce LociArt, a unified paradigm for free-style local video stylization that explicitly specifies where to stylize and what style to apply while preserving non-target content. At its core, GRNStyle builds on Generative Refinement Networks (GRN) and introduces a polar style-aware mask that embeds style identities as equal-norm directions within selected regions. Complementary representation--geometry conditioning combines encoded mask features with raw-mask spatial support, integrating spatiotemporal context with geometric control. To provide paired supervision, we develop an automated pipeline combining global stylization, video object removal, matting, and compositing, and construct the LociArt-100K Dataset with 100K local stylization pairs. We further establish the LociArt Benchmark, comprising 80 source videos and 84 target masks covering 168 foreground/background stylization cases, with a Style Prototype Similarity (SPS) evaluation scheme for regional style quality and complementary measures of non-target preservation and temporal consistency. Experiments show that GRNStyle outperforms existing methods in regional style quality and spatial control while maintaining strong temporal consistency. Qualitative examples further illustrate joint stylization of multiple regions with the same or different learned styles, despite training only on single-region, single-style pairs. Together, these components provide a unified foundation for learning and evaluating controllable local video stylization. Project page: https://higherhu.github.io/lociart/.

Tue 27 OctComputer Vision and Pattern RecognitionArtificial IntelligenceMultimedia
The gist
Making videos look like paintings or other art styles usually changes the whole scene, but sometimes you want to change just parts of the video. The authors introduce a method called LociArt that lets you pick exactly where and how to change the look while keeping the rest untouched. They also created a large dataset and evaluation tools to help computers learn and judge how well this selective styling works over time. Their approach shows it can apply styles smoothly to different parts even in moving videos.
Open → 2610.36456v1

Layout generation adapts progressively to current visual design state

PLAYGROUND: Progressive Layout Generation with Render-Grounded Planning

Abstract: Generating editable graphic designs requires arranging visual and textual elements into a coherent composition while preserving their content and editability. Although recent MLLM-based approaches employ iterative generation and visual refinement to reflect the real-world design workflows, their high-level plans are predetermined prior to execution, limiting their adaptability as the design evolves. To address this limitation, we propose PlayGround, a progressive layout generation framework that dynamically adapts to the current rendered canvas. PlayGround establishes a progressive design loop comprising a Progressive Planner, a Layout Generator, and a Visual Verifier, guided by a Structure-Aware Retriever. Given an initial set of unordered elements, the Structure-Aware Retriever initially infers a coarse layout structure to retrieve structurally similar reference designs. Subsequently, at each turn, the Progressive Planner examines the current rendered canvas alongside the remaining elements, forms the next functional group, and plans its composition. The Layout Generator then predicts editable layout attributes, while the Visual Verifier evaluates rendered candidates and provides feedback when refinement is necessary. To train the Progressive Planner with step-wise supervision, we construct a two-stage data generation pipeline that extracts functional groups and turn-level planning trajectories from completed designs. Experiments on Crello and LICA demonstrate the effectiveness of PlayGround over existing methods. Our code and weights will be publicly released.

Tue 13 OctComputer Vision and Pattern Recognition
The gist
Making graphic designs that can be easily changed means putting visual pieces in a good order while keeping them editable. The authors created PlayGround, a method that adjusts its design plan step-by-step based on what the design looks like at each stage, instead of planning everything from the start. It uses parts that plan the next group of elements, generate the layout, and check the visuals, improving the design progressively. Their tests show PlayGround works better than previous methods on popular design datasets.
Open → 2610.18865v1

Adaptive attention speeds and improves high resolution image generation

AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

Abstract: Generating images beyond the training resolution of Diffusion Transformers (DiTs) remains challenging, as the number of tokens grows rapidly with resolution, full attention becomes expensive, and outputs often suffer from structural distortions and degraded detail. We identify uniform full attention as the common source of both problems. Attending to all tokens at every step causes quadratic cost and exposes the model to more tokens and relative distances beyond the training range, pushing the model out of distribution. We observe that attention's effective receptive field narrows during denoising, from global in early steps that form structure to local in later steps that refine details. We propose AReA, a training-free attention mechanism whose receptive field evolves from coarse to fine, following the denoising progression. In early steps, AReA applies cross-scale coarse attention over downsampled keys and values, capturing global context, while in later steps it applies local attention within progressively shrinking windows. Without additional parameters or training, experiments show that AReA consistently outperforms state-of-the-art training-free baselines across target resolutions and different models, while also being faster and more efficient.

Mon 12 OctComputer Vision and Pattern Recognition
The gist
Making images larger than what a model was trained on is tricky because the attention system in the model must look at many more pieces of the image, which slows everything down and can lower quality. The authors found that paying attention to all parts equally at every step causes problems. They created AReA, a method that focuses on big-picture global attention early on, and then gradually narrows to smaller, local regions as the image is refined. This changes the model's focus in a smart way without retraining, making high-resolution image generation both faster and better.
Open → 2610.17227v1

Improving image generation by guiding pixel diffusion transformers

Learning Representations to Guide Pixel Diffusion Transformers

Abstract: Visual representations are largely developed for understanding, yet what makes them useful for generation remains unclear. We study this question in pixel diffusion transformers and uncover two key insights. First, better representation quality does not necessarily translate into greater generative utility, as externally aligned semantics can be progressively lost during pixel synthesis. Second, architectural separation with specialized pixel decoders mainly redistributes semantic and spatial modeling, without necessarily yielding representations better suited to generation. These insights suggest that a useful generative representation must be both properly formed and actively exploited. Motivated by these insights, we introduce Representation Guidance (RG), which directly forms and exploits internal representations while retaining a plain Transformer architecture. A low-dimensional bottleneck forces generation-relevant information into a compact representation, which is then explicitly used to guide decoder generation by modifying the flow-matching target according to its induced prediction offset. RG learns clearer semantic and spatial structure while scaling efficiently in both parameters and computation. On ImageNet 256x256, RG-L/16 with only 459M parameters outperforms substantially larger pixel diffusion transformers, while RG-XL/16 achieves an FID of 1.55. RG further reaches an FID of 1.74 at 512x512 and extends to text-to-image generation, achieving 0.89 on GenEval and 87.1 on DPG-Bench.

Fri 9 OctComputer Vision and Pattern Recognition
The gist
Generating images with computers involves turning abstract ideas into pixels. The authors found that making image representations better for understanding doesn’t always help when creating images, because some important details get lost along the way. They propose a new way called Representation Guidance that shapes and uses internal image information better during image creation. This method helps the system generate clearer and more accurate images, even when scaling up in size or adding text descriptions.
Open → 2610.12978v1

ProgressNet enables fast progressive image drawing with frozen models

ProgressNet: Sketching and Prompting with a Frozen Text-to-Image Model

Abstract: Humans draw progressively: a few strokes, a look at the result, a stroke erased, a prompt revised. Image generators do not work this way. They typically take a finished sketch and produce the image in a single pass, so every edit starts the picture again, and the models that do keep state across turns are driven by text, cannot take a stroke, and are too slow to draw with. We present ProgressNet, a training-free framework that lets a frozen text-to-image model follow a drawing session as it unfolds: strokes are added and erased, the prompt is revised, and the image keeps up at about a second per turn. It needs no new parameters because the frozen model already has what a progressive generator needs, a pathway through which the previous turn can be remembered, layers that can carry appearance forward without freezing structure, and an internal signal of how far to trust an unfinished sketch; three inference-time mechanisms (Previous-Concept Memory, Layer-Selective K/V Injection and Banded Adaptive Control) use each in turn. As a sketch fills in, every existing method degrades, the FID of the FLUX+ControlNet baseline doubling between 10% and 100% completion on FS-COCO, while ProgressNet's barely moves; it maintains strong fidelity and progressive coherence across three sketch domains and is preferred by users over five competitors, most widely on erasure.

Fri 2 OctComputer Vision and Pattern Recognition
The gist
Drawing pictures is something people do step by step—adding and erasing strokes as they go—but most AI image generators create images all at once and can’t easily update with new strokes or prompt changes. The authors developed ProgressNet, a method that lets a fixed text-to-image AI model keep up with ongoing drawing changes like adding or erasing strokes and revising text prompts, updating the image quickly each time. Their method doesn’t need to retrain the AI and keeps the image quality high throughout the drawing process, even as sketches become more complex.
Open → 2610.03512v1

Visual problem solving benchmark improves generative model planning

SolveEdit: Benchmarking Visual Problem Solving in Generative Models

Abstract: Machine intelligence is often evaluated through abstract reasoning problems, yet many real-world problems are visual, such as arranging objects, repairing layouts, or tracing routes. Solving these problems requires understanding a scene, inferring what must change to achieve a goal, and realizing that change without disturbing unrelated content. However, existing benchmarks mainly evaluate perception, generation, or explicitly specified transformations, leaving goal-driven visual problem solving underexplored. To bridge this gap, we introduce SolveEpIT, a benchmark for visual problem solving through scene transformation. Given an image and a goal, a model must infer a valid transformation from the request, the scene, or a visually expressed rule, then execute it while preserving unrelated content. SoLvEEDrr contains 2,728 cases. Atomic transition contracts specify required and protected conditions, enabling SoLvEScoRE to measure completion and unintended changes without a single reference output. The strongest evaluated model achieves only57.0% SolvEScore. We further introduce SolveEdiT-PLAN, a two-stage visual planner that instantiates the transition before generation. Under matched single-generation evaluation, it improves SoLvEScoRE by 9.1 points on average across three tested generators, including a gain from 57.0% to 71.6% for GPT-Image-2, without modifying the editor.

Mon 28 SeptComputer Vision and Pattern RecognitionArtificial Intelligence
The gist
Many real-world problems involve changing images to reach a goal, like moving objects without messing up the rest of the picture. The authors created a test called SolveEpIT to check if computer programs can do these tasks well. They also made a new two-step method that helps programs plan changes before making them, which works better than previous ways. Even the best current program only solved about half the problems correctly, showing this is a hard challenge.
Open → 2609.35504v1

Video text editing improved by aligning glyphs with text movement

Enhanced Video Text Editing with Trajectory-Aligned Glyph Rendering

Abstract: Video text editing aims to replace or add text in a video while keeping the rest of the video unchanged, which requires the edited text to be correct in every frame and to move coherently with the scene. Despite the remarkable progress of video diffusion models, they struggle to reproduce exact stroke structures and often produce garbled or wrong characters, especially for characters with complex strokes. To address this, we propose a trajectory-aligned glyph rendering reference that provides explicit per-frame glyph guidance following the position and perspective of the text, and a depth-normalized recognizer feature supervision that supervises the generated text on multi-depth features of a frozen text recognizer with per-depth normalized errors, targeting stroke errors overlooked by the diffusion loss. We further build VTEdit, a benchmark of 288 real-scene clips with 440 annotated text trajectories covering text replacement and text addition, which will be publicly released to facilitate future research. Experiments on VTEdit show that our method outperforms image text editing methods, video editing methods, and commercial models in text accuracy and background preservation, achieving a sentence accuracy of 0.9408, and receives the highest preference in a user study.

Mon 28 SeptComputer Vision and Pattern Recognition
The gist
Editing text in videos is hard because the new text needs to match the movement and angle of the original in every frame, which existing video diffusion methods struggle with. The authors created a way to guide the editing using exact shapes of letters matched to the text’s path and a new method to check mistakes in letter strokes more carefully. They also made a new set of 288 video clips with annotated text to test such methods. Their approach makes text changes more accurate and keeps the background better than previous methods and commercial tools.
Open → 2609.34178v1

Ruler improves svg image generation with detailed rubric rewards

RULER: Instance-aware Rubric Rewards for SVG Generation

Abstract: Generating Scalable Vector Graphics (SVG) code from natural-language instructions is an open-ended task without absolute visual ground truth, leaving both evaluation and policy optimization without a faithful signal. Scalar metrics (CLIP, Aesthetic) calibrated on natural images transfer poorly to stylized vector content, and reusing them as RL rewards triggers reward hacking. We address both limitations with rubric-based scoring. We first establish empirically that prompting a vision-language judge with a multi-axis rubric correlates with human judgments far better than scalar metrics, both across samples and within instructions. Building on this finding, we introduce RULER (Instance-aware Rubric Rewards for Reinforcement Learning), which converts each instruction into an instance-aware rubric of six items spanning semantic, visual, and stylistic axes; a judge VLM scores rendered rollouts item-by-item, and the weighted satisfactions form a fine-grained reward optimized via Group Relative Policy Optimization. Because the rubric is derived from text alone, RULER requires neither paired SVG ground truth nor human preference labels. On MMSVG-Illustration and MMSVG-Icon, RULER lifts the rubric score from 0.432/0.395 to 0.693/0.683, surpassing dedicated SVG specialists and matching the substantially larger DeepSeek-V3, with ablations identifying rubric design as the active lever for RL on open-ended SVG generation. The project page is available at https://hangyuran.github.io/RULER/.

Mon 21 SeptComputer Vision and Pattern Recognition
The gist
Generating SVG images from text instructions is tricky because there often isn’t one right answer, making it hard to evaluate or improve models. The authors show that using a detailed rubric with multiple criteria works better than simple scores for judging image quality. They created RULER, a system that makes a custom rubric for each instruction and scores generated images on various aspects using a vision-language model, then improves the generation through reinforcement learning based on these scores. This approach improves quality without needing exact example images or human feedback.
Open → 2609.25270v1

Model decodes graphic design elements in order to improve detection

Detect Anything in Graphic Design: Element-Level Rewards for Autoregressive Detection

Abstract: Graphic designs, such as posters, advertisements, and infographics, are an important medium for communicating information and shaping understanding. Unlike natural images, they consist of layered elements with explicit compositional order. However, existing object detection models treat these elements as an unordered set, leaving compositional order unexploited. To address this limitation, we present Detect Anything in Graphic Design (DAD), a model that formulates graphic design detection as compositional deconstruction. It decodes elements in compositional order, using lower-layer elements to better detect higher-layer ones. The key feature of DAD is amodal detection, which predicts the full bounding box of each element, including regions occluded by elements placed above it. Building on this formulation, we propose Element Relative Policy Optimization (EleRPO), which extends GRPO from sequence-level supervision to element-level optimization. EleRPO provides fine-grained training signals that capture how each detected element contributes to overall detection quality, and works synergistically with compositional order to improve detection performance. To support training and evaluation, we build a dataset of 10 million graphic designs. Experiments show that DAD outperforms all baselines and achieves human-level performance in amodal detection, supporting effective image-to-layer decomposition. EleRPO consistently improves over GRPO across nine detection benchmarks.

Mon 7 SeptComputer Vision and Pattern RecognitionMachine Learning
The gist
Graphic designs like posters are made of many layers stacked on top of each other, but current detection tools treat these layers like a jumbled set. The authors created a model, DAD, that looks at elements in the order they appear, using information from lower layers to detect higher ones better. Their system can also predict the full shape of each element, even if part of it is hidden behind others. They also developed a new training method that gives detailed feedback for each element to help the model learn better. They tested their approach on a huge dataset of graphic designs and found it works as well as humans for detecting covered elements.
Open → 2609.07072v1