arXiv subjects

Yi-Zhe Song

Publications and source records attributed to Yi-Zhe Song.

At least 19 recordsLinked to original sources

arXiv2026

Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation

Interactive autoregressive video generation demands both low-latency rollouts and precise online control. Few-step distillation accelerates generation by reducing denoising steps, while online control imposes a causal constraint: frames and blocks should depend on history and controls available during generation. Existing video distribution matching distillation (DMD) pipelines, however, often supervise causal few-step students using bidirectional teachers that score complete clips. The score for a target can therefore depend on future frames and controls that were unavailable when the student generated it, misaligning teacher supervision with the student's causal information set. We introduce Context-Matched Distillation (CMD), a causal DMD framework that aligns teacher supervision with the information available when each target is generated. CMD replaces bidirectional full-clip scoring with a causal teacher that evaluates each target without access to future frames or controls. The same causal teacher initializes the few-step student, establishing a consistent causal formulation across teacher training, student distillation, and inference. Beyond aligning the temporal information boundary, Prefix Scoring matches supervision to the student's realized rollout context by evaluating each target under the cached student-generated prefix that produced it. Prefix Corruption further stabilizes training by perturbing unreliable prefixes produced early in training while preserving this target-context alignment. With a simple causal formulation, CMD naturally extends to frame-wise and chunk-wise generation, long video distillation, and camera-conditioned distillation. Experiments demonstrate state-of-the-art aggregate performance among autoregressive methods on both short- and long-video benchmarks, together with substantially improved adherence to time-varying camera controls.

cs.CV↗

arXiv2026

Geometrically Consistent Multi-View Scene Generation from Freehand Sketches

We tackle a new problem: generating geometrically consistent multi-view scenes from a single freehand sketch. Freehand sketches are the most geometrically impoverished input one could offer a multi-view generator. They convey scene intent through abstract strokes while introducing spatial distortions that actively conflict with any consistent 3D interpretation. No prior method attempts this; existing multi-view approaches require photographs or text, while sketch-to-3D methods need multiple views or costly per-scene optimisation. We address three compounding challenges; absent training data, the need for geometric reasoning from distorted 2D input, and cross-view consistency, through three mutually reinforcing contributions: (i) a curated dataset of $\sim$9k sketch-to-multiview samples, constructed via an automated generation and filtering pipeline; (ii) Parallel Camera-Aware Attention Adapters (CA3) that inject geometric inductive biases into the video transformer; and (iii) a Sparse Correspondence Supervision Loss (CSL) derived from Structure-from-Motion reconstructions. Our framework synthesizes all views in a single denoising process without requiring reference images, iterative refinement, or per-scene optimization. Our approach significantly outperforms state-of-the-art two-stage baselines, improving realism (FID) by over 60% and geometric consistency (Corr-Acc) by 23%, while providing up to a 3.7$\times$ inference speedup.

cs.CV↗

arXiv2026

CaricHarmony: Contrastive Diffusion Paths for Identity-Preserving Caricature Synthesis

Sketch-based caricature synthesis suffers from a fundamental failure mode: when identity and shape conditions are combined in diffusion models, they create destructive interference that causes inevitable collapse toward either bland portraits or unrecognizable distortions. We identify the root cause as \emph{condition signal contamination} -- competing probability distributions in the denoising trajectory that make balanced generation impossible. We present CaricHarmony, the first training-free method that explicitly resolves this contamination through parallel uncontaminated diffusion paths. During inference, we maintain three paths: $\mathcal{P}^{\mathrm{i}}$ (pure identity), $\mathcal{P}^{\mathrm{s}}$ (pure shape), and $\mathcal{P}^{\mathrm{i+s}}$ (harmonized output). Novel energy functions operating on cross-attention features provide gradient guidance that steers $\mathcal{P}^{\mathrm{i+s}}$ toward optimal balance: $\mathcal{E}_{\mathrm{shape}}$ ensures sketch fidelity through layout and semantic alignment, while $\mathcal{E}_{\mathrm{id}}$ employs token-level correspondence matching robust to extreme distortions. Unlike DemoCaricature requiring 70 seconds per-identity fine-tuning or CaricatureBooth constrained to Bezier curves, CaricHarmony accepts any sketch format and generates in under 16 seconds. Experiments demonstrate state-of-the-art performance: 0.8615 shape CLIP score (vs. 0.8450) under comparable identity consistency score, with 7.81 overall user preference score (vs. 6.06). Our method fundamentally reconceptualizes the ID-shape conflict as conditioning signal contamination for diffusion models, enabling unprecedented creative control while preserving recognition.

cs.CV↗

arXiv2026

Chirpy3D: Part-Aware Multi-View Diffusion for Creative Fine-Grained Object Generation

Understanding and generating the fine-grained structure of objects -- such as birds with species-specific beaks, wings, and tails -- is a long-standing challenge in computer vision. We propose Chirpy3D, a part-aware multi-view diffusion framework that learns a hierarchical part latent space from unposed 2D images, using only off-the-shelf 2D part segmentation masks as spatial guidance -- without requiring any 3D data, camera poses, or manual part annotations. This latent space enables intuitive part-level swapping, interpolation, and zero-shot composition. A self-supervised feature consistency loss further encourages structural alignment across views, allowing coherent generation even with hybrid or unseen part combinations. Our core contribution is the controllable part-aware latent space and multi-view diffusion model. Downstream 3D generation is supported via any differentiable renderer such as NeRF but is orthogonal to the main framework, making Chirpy3D a flexible foundation for creative object generation in the absence of structured 3D data. Code is released at https://github.com/kamwoh/chirpy3d.

cs.CV↗

arXiv2026

SketchDeco: Training-Free Latent Composition for Precise Sketch Colourisation

We introduce SketchDeco, a training-free approach to sketch colourisation that bridges the gap between professional design needs and intuitive, region-based control. Our method empowers artists to use simple masks and colour palettes for precise spatial and chromatic specification, avoiding both the tediousness of manual assignment and the ambiguity of text-based prompts. We reformulate this task as a novel, training-free composition problem. Our core technical contribution is a guided latent-space blending process: we first leverage diffusion inversion to precisely ``paint'' user-defined colours into specified regions, and then use a custom self-attention mechanism to harmoniously blend these local edits with a globally consistent base image. This ensures both local colour fidelity and global harmony without requiring any model fine-tuning. Our system produces high-quality results in 15--20 inference steps on consumer GPUs, making professional-quality, controllable colourisation accessible.

cs.CV↗

arXiv2026

PerfGuard: A Performance-Aware Agent for Visual Content Generation

The advancement of Large Language Model (LLM)-powered agents has enabled automated task processing through reasoning and tool invocation capabilities. However, existing frameworks often operate under the idealized assumption that tool executions are invariably successful, relying solely on textual descriptions that fail to distinguish precise performance boundaries and cannot adapt to iterative tool updates. This gap introduces uncertainty in planning and execution, particularly in domains like visual content generation (AIGC), where nuanced tool performance significantly impacts outcomes. To address this, we propose PerfGuard, a performance-aware agent framework for visual content generation that systematically models tool performance boundaries and integrates them into task planning and scheduling. Our framework introduces three core mechanisms: (1) Performance-Aware Selection Modeling (PASM), which replaces generic tool descriptions with a multi-dimensional scoring system based on fine-grained performance evaluations; (2) Adaptive Preference Update (APU), which dynamically optimizes tool selection by comparing theoretical rankings with actual execution rankings; and (3) Capability-Aligned Planning Optimization (CAPO), which guides the planner to generate subtasks aligned with performance-aware strategies. Experimental comparisons against state-of-the-art methods demonstrate PerfGuard's advantages in tool selection accuracy, execution reliability, and alignment with user intent, validating its robustness and practical utility for complex AIGC tasks. The project code is available at https://github.com/FelixChan9527/PerfGuard.

cs.AI↗

arXiv2026

SynMind: Reducing Semantic Hallucination in fMRI-Based Image Reconstruction

Recent advances in fMRI-based image reconstruction have achieved remarkable photo-realistic fidelity. Yet, a persistent limitation remains: while reconstructed images often appear naturalistic and holistically similar to the target stimuli, they frequently suffer from severe semantic misalignment -- salient objects are often replaced or hallucinated despite high visual quality. In this work, we address this limitation by rethinking the role of explicit semantic interpretation in fMRI decoding. We argue that existing methods rely too heavily on entangled visual embeddings which prioritize low-level appearance cues -- such as texture and global gist -- over explicit semantic identity. To overcome this, we parse fMRI signals into rich, sentence-level semantic descriptions that mirror the hierarchical and compositional nature of human visual understanding. We achieve this by leveraging grounded VLMs to generate synthetic, human-like, multi-granularity textual representations that capture object identities and spatial organization. Built upon this foundation, we propose SynMind, a framework that integrates these explicit semantic encodings with visual priors to condition a pretrained diffusion model. Extensive experiments demonstrate that SynMind outperforms state-of-the-art methods across most quantitative metrics. Notably, by offloading semantic reasoning to our text-alignment module, SynMind surpasses competing methods based on SDXL while using the much smaller Stable Diffusion 1.4 and a single consumer GPU. Large-scale human evaluations further confirm that SynMind produces reconstructions more consistent with human visual perception. Neurovisualization analyses reveal that SynMind engages broader and more semantically relevant brain regions, mitigating the over-reliance on high-level visual areas.

cs.CV↗

arXiv2025

Block Cascading: Training Free Acceleration of Block-Causal Video Models

Block-causal video generation faces a stark speed-quality trade-off: small 1.3B models manage only 16 FPS while large 14B models crawl at 4.5 FPS, forcing users to choose between responsiveness and quality. Block Cascading significantly mitigates this trade-off through training-free parallelization. Our key insight: future video blocks do not need fully denoised current blocks to begin generation. By starting block generation with partially denoised context from predecessors, we transform sequential pipelines into parallel cascades where multiple blocks denoise simultaneously. With 5 GPUs exploiting temporal parallelism, we achieve ~2x acceleration across all model scales: 1.3B models accelerate from 16 to 30 FPS, 14B models from 4.5 to 12.5 FPS. Beyond inference speed, Block Cascading eliminates overhead from KV-recaching (of ~200ms) during context switches for interactive generation. Extensive evaluations validated against multiple block-causal pipelines demonstrate no significant loss in generation quality when switching from block-causal to Block Cascading pipelines for inference. Project Page: https://hmrishavbandy.github.io/block_cascading_page/

cs.CV↗

arXiv2025

4-Doodle: Text to 3D Sketches that Move!

We present a novel task: text-to-3D sketch animation, which aims to bring freeform sketches to life in dynamic 3D space. Unlike prior works focused on photorealistic content generation, we target sparse, stylized, and view-consistent 3D vector sketches, a lightweight and interpretable medium well-suited for visual communication and prototyping. However, this task is very challenging: (i) no paired dataset exists for text and 3D (or 4D) sketches; (ii) sketches require structural abstraction that is difficult to model with conventional 3D representations like NeRFs or point clouds; and (iii) animating such sketches demands temporal coherence and multi-view consistency, which current pipelines do not address. Therefore, we propose 4-Doodle, the first training-free framework for generating dynamic 3D sketches from text. It leverages pretrained image and video diffusion models through a dual-space distillation scheme: one space captures multi-view-consistent geometry using differentiable Bézier curves, while the other encodes motion dynamics via temporally-aware priors. Unlike prior work (e.g., DreamFusion), which optimizes from a single view per step, our multi-view optimization ensures structural alignment and avoids view ambiguity, critical for sparse sketches. Furthermore, we introduce a structure-aware motion module that separates shape-preserving trajectories from deformation-aware changes, enabling expressive motion such as flipping, rotation, and articulated movement. Extensive experiments show that our method produces temporally realistic and structurally stable 3D sketch animations, outperforming existing baselines in both fidelity and controllability. We hope this work serves as a step toward more intuitive and accessible 4D content creation.

cs.GR↗

arXiv2025

SD3.5-Flash: Distribution-Guided Distillation of Generative Flows

We present SD3.5-Flash, an efficient few-step distillation framework that brings high-quality image generation to accessible consumer devices. Our approach distills computationally prohibitive rectified flow models through a reformulated distribution matching objective tailored specifically for few-step generation. We introduce two key innovations: "timestep sharing" to reduce gradient noise and "split-timestep fine-tuning" to improve prompt alignment. Combined with comprehensive pipeline optimizations like text encoder restructuring and specialized quantization, our system enables both rapid generation and memory-efficient deployment across different hardware configurations. This democratizes access across the full spectrum of devices, from mobile phones to desktop computers. Through extensive evaluation including large-scale user studies, we demonstrate that SD3.5-Flash consistently outperforms existing few-step methods, making advanced generative AI truly accessible for practical deployment.

cs.CV↗

arXiv2025

Amadeus: Autoregressive Model with Bidirectional Attribute Modelling for Symbolic Music

Existing state-of-the-art symbolic music generation models predominantly adopt autoregressive or hierarchical autoregressive architectures, modelling symbolic music as a sequence of attribute tokens with unidirectional temporal dependencies, under the assumption of a fixed, strict dependency structure among these attributes. However, we observe that using different attributes as the initial token in these models leads to comparable performance. This suggests that the attributes of a musical note are, in essence, a concurrent and unordered set, rather than a temporally dependent sequence. Based on this insight, we introduce Amadeus, a novel symbolic music generation framework. Amadeus adopts a two-level architecture: an autoregressive model for note sequences and a bidirectional discrete diffusion model for attributes. To enhance performance, we propose Music Latent Space Discriminability Enhancement Strategy(MLSDES), incorporating contrastive learning constraints that amplify discriminability of intermediate music representations. The Conditional Information Enhancement Module (CIEM) simultaneously strengthens note latent vector representation via attention mechanisms, enabling more precise note decoding. We conduct extensive experiments on unconditional and text-conditioned generation tasks. Amadeus significantly outperforms SOTA models across multiple metrics while achieving at least 4$\times$ speed-up. Furthermore, we demonstrate training-free, fine-grained note attribute control feasibility using our model. To explore the upper performance bound of the Amadeus architecture, we compile the largest open-source symbolic music dataset to date, AMD (Amadeus MIDI Dataset), supporting both pre-training and fine-tuning.

cs.SD↗

arXiv2025

SketchAnimator: Animate Sketch via Motion Customization of Text-to-Video Diffusion Models

Sketching is a uniquely human tool for expressing ideas and creativity. The animation of sketches infuses life into these static drawings, opening a new dimension for designers. Animating sketches is a time-consuming process that demands professional skills and extensive experience, often proving daunting for amateurs. In this paper, we propose a novel sketch animation model SketchAnimator, which enables adding creative motion to a given sketch, like "a jumping car''. Namely, given an input sketch and a reference video, we divide the sketch animation into three stages: Appearance Learning, Motion Learning and Video Prior Distillation. In stages 1 and 2, we utilize LoRA to integrate sketch appearance information and motion dynamics from the reference video into the pre-trained T2V model. In the third stage, we utilize Score Distillation Sampling (SDS) to update the parameters of the Bezier curves in each sketch frame according to the acquired motion information. Consequently, our model produces a sketch video that not only retains the original appearance of the sketch but also mirrors the dynamic movements of the reference video. We compare our method with alternative approaches and demonstrate that it generates the desired sketch video under the challenge of one-shot motion customization.

cs.CV↗

arXiv2025

Annotation-Free Human Sketch Quality Assessment

As lovely as bunnies are, your sketched version would probably not do them justice (Fig.~\ref{fig:intro}). This paper recognises this very problem and studies sketch quality assessment for the first time -- letting you find these badly drawn ones. Our key discovery lies in exploiting the magnitude ($L_2$ norm) of a sketch feature as a quantitative quality metric. We propose Geometry-Aware Classification Layer (GACL), a generic method that makes feature-magnitude-as-quality-metric possible and importantly does it without the need for specific quality annotations from humans. GACL sees feature magnitude and recognisability learning as a dual task, which can be simultaneously optimised under a neat cross-entropy classification loss with theoretic guarantee. This gives GACL a nice geometric interpretation (the better the quality, the easier the recognition), and makes it agnostic to both network architecture changes and the underlying sketch representation. Through a large scale human study of 160,000 \doublecheck{trials}, we confirm the agreement between our GACL-induced metric and human quality perception. We further demonstrate how such a quality assessment capability can for the first time enable three practical sketch applications. Interestingly, we show GACL not only works on abstract visual representations such as sketch but also extends well to natural images on the problem of image quality assessment (IQA). Last but not least, we spell out the general properties of GACL as general-purpose data re-weighting strategy and demonstrate its applications in vertical problems such as noisy label cleansing. Code will be made publicly available at github.com/yanglan0225/SketchX-Quantifying-Sketch-Quality.

cs.CV↗

arXiv2025

Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection

Keypoint detection, integral to modern machine perception, faces challenges in few-shot learning, particularly when source data from the same distribution as the query is unavailable. This gap is addressed by leveraging sketches, a popular form of human expression, providing a source-free alternative. However, challenges arise in mastering cross-modal embeddings and handling user-specific sketch styles. Our proposed framework overcomes these hurdles with a prototypical setup, combined with a grid-based locator and prototypical domain adaptation. We also demonstrate success in few-shot convergence across novel keypoints and classes through extensive experiments.

cs.CV↗

arXiv2025

Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models

Negative guidance -- explicitly suppressing unwanted attributes -- remains a fundamental challenge in diffusion models, particularly in few-step sampling regimes. While Classifier-Free Guidance (CFG) works well in standard settings, it fails under aggressive sampling step compression due to divergent predictions between positive and negative branches. We present Normalized Attention Guidance (NAG), an efficient, training-free mechanism that applies extrapolation in attention space with L1-based normalization and refinement. NAG restores effective negative guidance where CFG collapses while maintaining fidelity. Unlike existing approaches, NAG generalizes across architectures (UNet, DiT), sampling regimes (few-step, multi-step), and modalities (image, video), functioning as a \textit{universal} plug-in with minimal computational overhead. Through extensive experimentation, we demonstrate consistent improvements in text alignment (CLIP Score), fidelity (FID, PFID), and human-perceived quality (ImageReward). Our ablation studies validate each design component, while user studies confirm significant preference for NAG-guided outputs. As a model-agnostic inference-time approach requiring no retraining, NAG provides effortless negative guidance for all modern diffusion frameworks -- pseudocode in the Appendix!

cs.CV↗

arXiv2025

Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch

As sketch research has collectively matured over time, its adaptation for at-mass commercialisation emerges on the immediate horizon. Despite an already mature research endeavour for photos, there is no research on the efficient inference specifically designed for sketch data. In this paper, we first demonstrate existing state-of-the-art efficient light-weight models designed for photos do not work on sketches. We then propose two sketch-specific components which work in a plug-n-play manner on any photo efficient network to adapt them to work on sketch data. We specifically chose fine-grained sketch-based image retrieval (FG-SBIR) as a demonstrator as the most recognised sketch problem with immediate commercial value. Technically speaking, we first propose a cross-modal knowledge distillation network to transfer existing photo efficient networks to be compatible with sketch, which brings down number of FLOPs and model parameters by 97.96% percent and 84.89% respectively. We then exploit the abstract trait of sketch to introduce a RL-based canvas selector that dynamically adjusts to the abstraction level which further cuts down number of FLOPs by two thirds. The end result is an overall reduction of 99.37% of FLOPs (from 40.18G to 0.254G) when compared with a full network, while retaining the accuracy (33.03% vs 32.77%) -- finally making an efficient network for the sparse sketch data that exhibit even fewer FLOPs than the best photo counterpart.

cs.CV↗

arXiv2025

SketchFusion: Learning Universal Sketch Features through Fusing Foundation Models

While foundation models have revolutionised computer vision, their effectiveness for sketch understanding remains limited by the unique challenges of abstract, sparse visual inputs. Through systematic analysis, we uncover two fundamental limitations: Stable Diffusion (SD) struggles to extract meaningful features from abstract sketches (unlike its success with photos), and exhibits a pronounced frequency-domain bias that suppresses essential low-frequency components needed for sketch understanding. Rather than costly retraining, we address these limitations by strategically combining SD with CLIP, whose strong semantic understanding naturally compensates for SD's spatial-frequency biases. By dynamically injecting CLIP features into SD's denoising process and adaptively aggregating features across semantic levels, our method achieves state-of-the-art performance in sketch retrieval (+3.35%), recognition (+1.06%), segmentation (+29.42%), and correspondence learning (+21.22%), demonstrating the first truly universal sketch feature representation in the era of foundation models.

cs.CV↗

arXiv2025

SketchYourSeg: Mask-Free Subjective Image Segmentation via Freehand Sketches

We introduce SketchYourSeg, a novel framework that establishes freehand sketches as a powerful query modality for subjective image segmentation across entire galleries through a single exemplar sketch. Unlike text prompts that struggle with spatial specificity or interactive methods confined to single-image operations, sketches naturally combine semantic intent with structural precision. This unique dual encoding enables precise visual disambiguation for segmentation tasks where text descriptions would be cumbersome or ambiguous -- such as distinguishing between visually similar instances, specifying exact part boundaries, or indicating spatial relationships in composed concepts. Our approach addresses three fundamental challenges: (i) eliminating the need for pixel-perfect annotation masks during training with a mask-free framework; (ii) creating a synergistic relationship between sketch-based image retrieval (SBIR) models and foundation models (CLIP/DINOv2) where the former provides training signals while the latter generates masks; and (iii) enabling multi-granular segmentation capabilities through purpose-made sketch augmentation strategies. Our extensive evaluations demonstrate superior performance over existing approaches across diverse benchmarks, establishing a new paradigm for user-guided image segmentation that balances precision with efficiency.

cs.CV↗