arXiv subjects

Yue Ma

Publications and source records attributed to Yue Ma.

At least 109 records · Page 6Linked to original sources

arXiv2025

SkipVAR: Accelerating Visual Autoregressive Modeling via Adaptive Frequency-Aware Skipping

Recent studies on Visual Autoregressive (VAR) models have highlighted that high-frequency components, or later steps, in the generation process contribute disproportionately to inference latency. However, the underlying computational redundancy involved in these steps has yet to be thoroughly investigated. In this paper, we conduct an in-depth analysis of the VAR inference process and identify two primary sources of inefficiency: step redundancy and unconditional branch redundancy. To address step redundancy, we propose an automatic step-skipping strategy that selectively omits unnecessary generation steps to improve efficiency. For unconditional branch redundancy, we observe that the information gap between the conditional and unconditional branches is minimal. Leveraging this insight, we introduce unconditional branch replacement, a technique that bypasses the unconditional branch to reduce computational cost. Notably, we observe that the effectiveness of acceleration strategies varies significantly across different samples. Motivated by this, we propose SkipVAR, a sample-adaptive framework that leverages frequency information to dynamically select the most suitable acceleration strategy for each instance. To evaluate the role of high-frequency information, we introduce high-variation benchmark datasets that test model sensitivity to fine details. Extensive experiments show SkipVAR achieves over 0.88 average SSIM with up to 1.81x overall acceleration and 2.62x speedup on the GenEval benchmark, maintaining model quality. These results confirm the effectiveness of frequency-aware, training-free adaptive acceleration for scalable autoregressive image generation. Our code is available at https://github.com/fakerone-li/SkipVAR and has been publicly released.

cs.CV↗

arXiv2025

Calligrapher: Freestyle Text Image Customization

We introduce Calligrapher, a novel diffusion-based framework that innovatively integrates advanced text customization with artistic typography for digital calligraphy and design applications. Addressing the challenges of precise style control and data dependency in typographic customization, our framework incorporates three key technical contributions. First, we develop a self-distillation mechanism that leverages the pre-trained text-to-image generative model itself alongside the large language model to automatically construct a style-centric typography benchmark. Second, we introduce a localized style injection framework via a trainable style encoder, which comprises both Qformer and linear layers, to extract robust style features from reference images. An in-context generation mechanism is also employed to directly embed reference images into the denoising process, further enhancing the refined alignment of target styles. Extensive quantitative and qualitative evaluations across diverse fonts and design contexts confirm Calligrapher's accurate reproduction of intricate stylistic details and precise glyph positioning. By automating high-quality, visually consistent typography, Calligrapher surpasses traditional models, empowering creative practitioners in digital art, branding, and contextual typographic design.

cs.CV↗

arXiv2025

Taming Rectified Flow for Inversion and Editing

Rectified-flow-based diffusion transformers like FLUX and OpenSora have demonstrated outstanding performance in the field of image and video generation. Despite their robust generative capabilities, these models often struggle with inversion inaccuracies, which could further limit their effectiveness in downstream tasks such as image and video editing. To address this issue, we propose RF-Solver, a novel training-free sampler that effectively enhances inversion precision by mitigating the errors in the ODE-solving process of rectified flow. Specifically, we derive the exact formulation of the rectified flow ODE and apply the high-order Taylor expansion to estimate its nonlinear components, significantly enhancing the precision of ODE solutions at each timestep. Building upon RF-Solver, we further propose RF-Edit, a general feature-sharing-based framework for image and video editing. By incorporating self-attention features from the inversion process into the editing process, RF-Edit effectively preserves the structural information of the source image or video while achieving high-quality editing results. Our approach is compatible with any pre-trained rectified-flow-based models for image and video tasks, requiring no additional training or optimization. Extensive experiments across generation, inversion, and editing tasks in both image and video modalities demonstrate the superiority and versatility of our method. The source code is available at https://github.com/wangjiangshan0725/RF-Solver-Edit.

cs.CV↗

arXiv2025

Follow-Your-Creation: Empowering 4D Creation through Video Inpainting

We introduce Follow-Your-Creation, a novel 4D video creation framework capable of both generating and editing 4D content from a single monocular video input. By leveraging a powerful video inpainting foundation model as a generative prior, we reformulate 4D video creation as a video inpainting task, enabling the model to fill in missing content caused by camera trajectory changes or user edits. To facilitate this, we generate composite masked inpainting video data to effectively fine-tune the model for 4D video generation. Given an input video and its associated camera trajectory, we first perform depth-based point cloud rendering to obtain invisibility masks that indicate the regions that should be completed. Simultaneously, editing masks are introduced to specify user-defined modifications, and these are combined with the invisibility masks to create a composite masks dataset. During training, we randomly sample different types of masks to construct diverse and challenging inpainting scenarios, enhancing the model's generalization and robustness in various 4D editing and generation tasks. To handle temporal consistency under large camera motion, we design a self-iterative tuning strategy that gradually increases the viewing angles during training, where the model is used to generate the next-stage training data after each fine-tuning iteration. Moreover, we introduce a temporal packaging module during inference to enhance generation quality. Our method effectively leverages the prior knowledge of the base model without degrading its original performance, enabling the generation of 4D videos with consistent multi-view coherence. In addition, our approach supports prompt-based content editing, demonstrating strong flexibility and significantly outperforming state-of-the-art methods in both quality and versatility.

cs.CV↗

arXiv2025

Generalized Turán problem with bounded matching number

For a graph $T$ and a set of graphs $\mathcal{H}$, let $\mbox{ex}(n,T,\mathcal{H})$ denote the maximum number of copies of $T$ in an $n$-vertex $\mathcal{H}$-free graph. Recently, Alon and Frankl~(arXiv2210.15076) determined the exact value of $\mbox{ex}(n,K_2,\{K_{k+1},M_{s+1}\})$, where $K_{k+1}$ and $M_{s+1}$ are complete graph on $k+1$ vertices and matching of size $s+1$, respectively. Soon after, Gerbner~(arXiv2211.03272) continued the study by extending $K_{k+1}$ to general fixed graph $H$. In this paper, we continue the study of the function $\mbox{ex}(n, T,\{H,M_{s+1}\})$ when $T=K_r$ for $r\ge 3$. We determine the exact value of $\mbox{ex}(n,K_r,\{K_{k+1},M_{s+1}\})$ and give the value of $\mbox{ex}(n,K_r,\{H,M_{s+1}\})$ for general $H$ with an error term $O(1)$.

math.CO↗

arXiv2025

Reshaping quantum device noise via repetition code circuits

Noise of a quantum processor can be an important resource for simulating open quantum dynamics. However, this requires characterizing the device noise and then transforming it into a target structure. Here we take the first step towards this goal: We analytically and numerically study reshaping the noise associated with native trapped-ion two-qubit entangling gates via quantum circuits based on repetition codes, and experimentally demonstrate our findings on the IonQ Aria-1 quantum hardware. We investigate all the building blocks, including the quantum channels describing noisy two-qubit entangling gates, the compilation of the encoding circuits into native gates, and the propagation of two-qubit errors across ideal single-qubit gates.

quant-ph↗

arXiv2025

MultiBooth: Towards Generating All Your Concepts in an Image from Text

This paper introduces MultiBooth, a novel and efficient technique for multi-concept customization in image generation from text. Despite the significant advancements in customized generation methods, particularly with the success of diffusion models, existing methods often struggle with multi-concept scenarios due to low concept fidelity and high inference cost. MultiBooth addresses these issues by dividing the multi-concept generation process into two phases: a single-concept learning phase and a multi-concept integration phase. During the single-concept learning phase, we employ a multi-modal image encoder and an efficient concept encoding technique to learn a concise and discriminative representation for each concept. In the multi-concept integration phase, we use bounding boxes to define the generation area for each concept within the cross-attention map. This method enables the creation of individual concepts within their specified regions, thereby facilitating the formation of multi-concept images. This strategy not only improves concept fidelity but also reduces additional inference cost. MultiBooth surpasses various baselines in both qualitative and quantitative evaluations, showcasing its superior performance and computational efficiency. Project Page: https://multibooth.github.io/

cs.CV↗

arXiv2025

AvatarArtist: Open-Domain 4D Avatarization

This work focuses on open-domain 4D avatarization, with the purpose of creating a 4D avatar from a portrait image in an arbitrary style. We select parametric triplanes as the intermediate 4D representation and propose a practical training paradigm that takes advantage of both generative adversarial networks (GANs) and diffusion models. Our design stems from the observation that 4D GANs excel at bridging images and triplanes without supervision yet usually face challenges in handling diverse data distributions. A robust 2D diffusion prior emerges as the solution, assisting the GAN in transferring its expertise across various domains. The synergy between these experts permits the construction of a multi-domain image-triplane dataset, which drives the development of a general 4D avatar creator. Extensive experiments suggest that our model, AvatarArtist, is capable of producing high-quality 4D avatars with strong robustness to various source image domains. The code, the data, and the models will be made publicly available to facilitate future studies.

cs.CV↗

arXiv2025

Minimum saturated graphs without $4$-cycles and $5$-cycles

Given a family of graphs $\mathcal{F}$, a graph $G$ is said to be $\mathcal{F}$-saturated if $G$ does not contain a copy of $F$ as a subgraph for any $F\in\mathcal{F}$, but the addition of any edge $e\notin E(G)$ creates at least one copy of some $F\in\mathcal{F}$ within $G$. The minimum size of an $\mathcal{F}$-saturated graph on $n$ vertices is called the saturation number, denoted by $\mbox{sat}(n, \mathcal{F})$. Let $C_r$ be the cycle of length $r$. In this paper, we study on $\mbox{sat}(n, \mathcal{F})$ when $\mathcal{F}$ is a family of cycles. In particular, we determine that $\mbox{sat}(n, \{C_4,C_5\})=\lceil\frac{5n}{4}-\frac{3}{2}\rceil$ for any positive integer $n$.

math.CO↗

arXiv2025

Towards Multiple Character Image Animation Through Enhancing Implicit Decoupling

Controllable character image animation has a wide range of applications. Although existing studies have consistently improved performance, challenges persist in the field of character image animation, particularly concerning stability in complex backgrounds and tasks involving multiple characters. To address these challenges, we propose a novel multi-condition guided framework for character image animation, employing several well-designed input modules to enhance the implicit decoupling capability of the model. First, the optical flow guider calculates the background optical flow map as guidance information, which enables the model to implicitly learn to decouple the background motion into background constants and background momentum during training, and generate a stable background by setting zero background momentum during inference. Second, the depth order guider calculates the order map of the characters, which transforms the depth information into the positional information of multiple characters. This facilitates the implicit learning of decoupling different characters, especially in accurately separating the occluded body parts of multiple characters. Third, the reference pose map is input to enhance the ability to decouple character texture and pose information in the reference image. Furthermore, to fill the gap of fair evaluation of multi-character image animation, we propose a new benchmark comprising about 4,000 frames. Extensive qualitative and quantitative evaluations demonstrate that our method excels in generating high-quality character animations, especially in scenarios of complex backgrounds and multiple characters.

cs.CV↗

arXiv2025

Cylindrical cavity expansion analysis under partially drained conditions for normalisation of excess water pressure in CPTU

Cone tip resistance and excess water pressure (EWP) measured by piezocone penetration tests (CPTU) may be significantly affected by the partially drained effect in soils with intermediate permeability. To capture this effect, the paper proposes a straightforward, hydro-mechanical coupling solution for cylindrical cavity expansion under partially drained conditions. The mechanical behaviour of soils is modelled using the elastoplastic Tresca model and water flow within porous soils is assumed to obey Darcys law. Two partial differential equations (PDEs) are established in the elastic and plastic zones, respectively, transforming the cavity expansion analysis into a typical Stefan problem with dynamic boundary conditions (i.e. a moving boundary at the elastoplastic interface). An approximate solution for the PDEs is derived by leveraging the variable transformation method. Based on the new solution, a novel normalised penetration rate is defined considering the rigidity index of soils, with which a unique backbone curve for CPTU is found. Finally, the backbone curve is compared with a database comprising 109 in-situ experimental tests, 101 centrifuge modelling tests, and numerical simulation results. The proposed solution may provide a useful theoretical tool for interpreting the consolidation coefficient of fine-grained soils from the penetration stage of multi-rate CPTU, which can enhance the interpretation reliability for CPTU dissipation tests.

physics.geo-ph↗

arXiv2025

Enhancing Image Generation Fidelity via Progressive Prompts

The diffusion transformer (DiT) architecture has attracted significant attention in image generation, achieving better fidelity, performance, and diversity. However, most existing DiT - based image generation methods focus on global - aware synthesis, and regional prompt control has been less explored. In this paper, we propose a coarse - to - fine generation pipeline for regional prompt - following generation. Specifically, we first utilize the powerful large language model (LLM) to generate both high - level descriptions of the image (such as content, topic, and objects) and low - level descriptions (such as details and style). Then, we explore the influence of cross - attention layers at different depths. We find that deeper layers are always responsible for high - level content control, while shallow layers handle low - level content control. Various prompts are injected into the proposed regional cross - attention control for coarse - to - fine generation. By using the proposed pipeline, we enhance the controllability of DiT - based image generation. Extensive quantitative and qualitative results show that our pipeline can improve the performance of the generated images.

cs.CV↗

arXiv2025

ListConRanker: A Contrastive Text Reranker with Listwise Encoding

Reranker models aim to re-rank the passages based on the semantics similarity between the given query and passages, which have recently received more attention due to the wide application of the Retrieval-Augmented Generation. Most previous methods apply pointwise encoding, meaning that it can only encode the context of the query for each passage input into the model. However, for the reranker model, given a query, the comparison results between passages are even more important, which is called listwise encoding. Besides, previous models are trained using the cross-entropy loss function, which leads to issues of unsmooth gradient changes during training and low training efficiency. To address these issues, we propose a novel Listwise-encoded Contrastive text reRanker (ListConRanker). It can help the passage to be compared with other passages during the encoding process, and enhance the contrastive information between positive examples and between positive and negative examples. At the same time, we use the circle loss to train the model to increase the flexibility of gradients and solve the problem of training efficiency. Experimental results show that ListConRanker achieves state-of-the-art performance on the reranking benchmark of Chinese Massive Text Embedding Benchmark, including the cMedQA1.0, cMedQA2.0, MMarcoReranking, and T2Reranking datasets.

cs.CL↗

arXiv2025

H-MBA: Hierarchical MamBa Adaptation for Multi-Modal Video Understanding in Autonomous Driving

With the prevalence of Multimodal Large Language Models(MLLMs), autonomous driving has encountered new opportunities and challenges. In particular, multi-modal video understanding is critical to interactively analyze what will happen in the procedure of autonomous driving. However, videos in such a dynamical scene that often contains complex spatial-temporal movements, which restricts the generalization capacity of the existing MLLMs in this field. To bridge the gap, we propose a novel Hierarchical Mamba Adaptation (H-MBA) framework to fit the complicated motion changes in autonomous driving videos. Specifically, our H-MBA consists of two distinct modules, including Context Mamba (C-Mamba) and Query Mamba (Q-Mamba). First, C-Mamba contains various types of structure state space models, which can effectively capture multi-granularity video context for different temporal resolutions. Second, Q-Mamba flexibly transforms the current frame as the learnable query, and attentively selects multi-granularity video context into query. Consequently, it can adaptively integrate all the video contexts of multi-scale temporal resolutions to enhance video understanding. Via a plug-and-play paradigm in MLLMs, our H-MBA shows the remarkable performance on multi-modal video tasks in autonomous driving, e.g., for risk object detection, it outperforms the previous SOTA method with 5.5% mIoU improvement.

cs.CV↗

arXiv2025

Penalized Quasi-likelihood for High-dimensional Longitudinal Data via Within-cluster Resampling

The generalized estimating equation (GEE) method is a popular tool for longitudinal data analysis. However, GEE produces biased estimates when the outcome of interest is associated with cluster size, a phenomenon known as informative cluster size (ICS). In this study, we address this issue by formulating the impact of ICS and proposing an integrated approach to mitigate its effects. Our method combines the concept of within-cluster resampling with a penalized quasi-likelihood framework applied to each resampled dataset, ensuring consistency in model selection and estimation. To aggregate the estimators from the resampled datasets, we introduce a penalized mean regression technique, resulting in a final estimator that improves true positive discovery rates while reducing false positives. Simulation studies and an application to yeast cell-cycle gene expression data demonstrate the excellent performance of the proposed penalized quasi-likelihood method via within-cluster resampling.

stat.ME↗

arXiv2025

Penalized weighted GEEs for high-dimensional longitudinal data with informative cluter size

High-dimensional longitudinal data have become increasingly prevalent in recent studies, and penalized generalized estimating equations (GEEs) are often used to model such data. However, the desirable properties of the GEE method can break down when the outcome of interest is associated with cluster size, a phenomenon known as informative cluster size. In this article, we address this issue by formulating the effect of informative cluster size and proposing a novel weighted GEE approach to mitigate its impact, while extending the penalized version for high-dimensional settings. We show that the penalized weighted GEE approach achieves consistency in both model selection and estimation. Theoretically, we establish that the proposed penalized weighted GEE estimator is asymptotically equivalent to the oracle estimator, assuming the true model is known. This result indicates that the penalized weighted GEE approach retains the excellent properties of the GEE method and is robust to informative cluster sizes, thereby extending its applicability to highly complex situations. Simulations and a real data application further demonstrate that the penalized weighted GEE outperforms the existing alternative methods.

stat.ME↗

arXiv2024

Follow-Your-MultiPose: Tuning-Free Multi-Character Text-to-Video Generation via Pose Guidance

Text-editable and pose-controllable character video generation is a challenging but prevailing topic with practical applications. However, existing approaches mainly focus on single-object video generation with pose guidance, ignoring the realistic situation that multi-character appear concurrently in a scenario. To tackle this, we propose a novel multi-character video generation framework in a tuning-free manner, which is based on the separated text and pose guidance. Specifically, we first extract character masks from the pose sequence to identify the spatial position for each generating character, and then single prompts for each character are obtained with LLMs for precise text guidance. Moreover, the spatial-aligned cross attention and multi-branch control module are proposed to generate fine grained controllable multi-character video. The visualized results of generating video demonstrate the precise controllability of our method for multi-character generation. We also verify the generality of our method by applying it to various personalized T2I models. Moreover, the quantitative results show that our approach achieves superior performance compared with previous works.

cs.CV↗

arXiv2024

COVE: Unleashing the Diffusion Feature Correspondence for Consistent Video Editing

Video editing is an emerging task, in which most current methods adopt the pre-trained text-to-image (T2I) diffusion model to edit the source video in a zero-shot manner. Despite extensive efforts, maintaining the temporal consistency of edited videos remains challenging due to the lack of temporal constraints in the regular T2I diffusion model. To address this issue, we propose COrrespondence-guided Video Editing (COVE), leveraging the inherent diffusion feature correspondence to achieve high-quality and consistent video editing. Specifically, we propose an efficient sliding-window-based strategy to calculate the similarity among tokens in the diffusion features of source videos, identifying the tokens with high correspondence across frames. During the inversion and denoising process, we sample the tokens in noisy latent based on the correspondence and then perform self-attention within them. To save GPU memory usage and accelerate the editing process, we further introduce the temporal-dimensional token merging strategy, which can effectively reduce redundancy. COVE can be seamlessly integrated into the pre-trained T2I diffusion model without the need for extra training or optimization. Extensive experiment results demonstrate that COVE achieves the start-of-the-art performance in various video editing scenarios, outperforming existing methods both quantitatively and qualitatively. The code will be release at https://github.com/wangjiangshan0725/COVE.

cs.CV↗