Searcharxiv⌕ Search

arXiv subjects

Shuai Li

Publications and source records attributed to Shuai Li.

At least 91 records · Page 5Linked to original sources
arXiv2026

A Noise Constrained Diffusion (NC-Diffusion) Framework for High Fidelity Image Compression

With the great success of diffusion models in image generation, diffusion-based image compression is attracting increasing interests. However, due to the random noise introduced in the diffusion learning, they usually produce reconstructions with deviation from the original images, leading to suboptimal compression results. To address this problem, in this paper, we propose a Noise Constrained Diffusion (NC-Diffusion) framework for high fidelity image compression. Unlike existing diffusion-based compression methods that add random Gaussian noise and direct the noise into the image space, the proposed NC-Diffusion formulates the quantization noise originally added in the learned image compression as the noise in the forward process of diffusion. Then a noise constrained diffusion process is constructed from the ground-truth image to the initial compression result generated with quantization noise. The NC-Diffusion overcomes the problem of noise mismatch between compression and diffusion, significantly improving the inference efficiency. In addition, an adaptive frequency-domain filtering module is developed to enhance the skip connections in the U-Net based diffusion architecture, in order to enhance high-frequency details. Moreover, a zero-shot sample-guided enhancement method is designed to further improve the fidelity of the image. Experiments on multiple benchmark datasets demonstrate that our method can achieve the best performance compared with existing methods.

eess.IV↗
arXiv2026

LiftFormer: Lifting and Frame Theory Based Monocular Depth Estimation Using Depth and Edge Oriented Subspace Representation

Monocular depth estimation (MDE) has attracted increasing interest in the past few years, owing to its important role in 3D vision. MDE is the estimation of a depth map from a monocular image/video to represent the 3D structure of a scene, which is a highly ill-posed problem. To solve this problem, in this paper, we propose a LiftFormer based on lifting theory topology, for constructing an intermediate subspace that bridges the image color features and depth values, and a subspace that enhances the depth prediction around edges. MDE is formulated by transforming the depth value prediction problem into depth-oriented geometric representation (DGR) subspace feature representation, thus bridging the learning from color values to geometric depth values. A DGR subspace is constructed based on frame theory by using linearly dependent vectors in accordance with depth bins to provide a redundant and robust representation. The image spatial features are transformed into the DGR subspace, where these features correspond directly to the depth values. Moreover, considering that edges usually present sharp changes in a depth map and tend to be erroneously predicted, an edge-aware representation (ER) subspace is constructed, where depth features are transformed and further used to enhance the local features around edges. The experimental results demonstrate that our LiftFormer achieves state-of-the-art performance on widely used datasets, and an ablation study validates the effectiveness of both proposed lifting modules in our LiftFormer.

cs.CV↗
arXiv2026

Cavitation-bubble Interaction with an Initially Perturbed Free Surface

The interaction of a spark-generated cavitation bubble with an initially perturbed free surface is investigated experimentally, numerically, and analytically. By exploiting contact-line pinning, we accurately prescribe an initial meniscus with a thin, hydrophilic-coated rod inserted into the liquid. A pronounced surface cavity, driven by the oscillating bubble, forms and penetrates downward to a scale comparable to the bubble itself. The coupled cavity-bubble system exhibits two distinct regimes -- coalescence and non-coalescence -- separated by a critical condition governed by the non-dimensional stand-off parameter $γ$ and the initial meniscus height $h_m$. In the non-coalescence regime, the cavity evolves through inception, expansion, and rebound/jetting. The maximum cavity length $h_c$ follows a power-law scaling $h_c\proptoγ^α$ with $α=-2.7$ (experiments) and $α=-2.6$ (simulations) for $1.5\lesssimγ\lesssim3$, where inertia dominates. Deviations emerge for $γ\lesssim1.5$ (strong nonlinearity) and $γ\gtrsim3$ (surface tension and viscosity become noticeable). An analytical model based on the Rayleigh-Plesset equation combined with nonlinear Rayleigh-Taylor instability theory captures the trend and confirms that $h_m$ plays only a secondary role relative to $γ$. In the coalescence regime, atmospheric air vents into the bubble through the merged cavity, weakening the collapse intensity and reducing the associated pressure peak. We also examine air/liquid compressibility and boundary layer effects, whose significance grows as $γ$ decreases. These findings are relevant to surface-jetting technologies, cavitation-erosion mitigation, and underwater-noise suppression.

physics.flu-dyn↗
arXiv2026

Determined by User Needs: A Salient Object Detection Rationale Beyond Conventional Visual Stimuli

Existing \textbf{s}alient \textbf{o}bject \textbf{d}etection (SOD) methods adopt a \textbf{passive} visual stimulus-based rationale--objects with the strongest visual stimuli are perceived as the user's primary focus (i.e., salient objects). They ignore the decisive role of users' \textbf{proactive needs} in segmenting salient objects--if a user has a need before seeing an image, the user's salient objects align with their needs, e.g., if a user's need is ``white apple'', when this user sees an image, the user's primary focus is on the ``white apple'' or ``the most white apple-like'' objects in the image. Such an oversight not only \textbf{fails to satisfy users}, but also \textbf{limits the development of downstream tasks}. For instance, in salient object ranking tasks, focusing solely on visual stimuli-based salient objects is insufficient for conducting an analysis of fine-grained relationships between users' viewing order (usually determined by user's needs) and scenes, which may result in wrong ranking results. Clearly, it is essential to detect salient objects based on user needs. Thus, we advocate a \textbf{User} \textbf{S}alient \textbf{O}bject \textbf{D}etection (UserSOD) task, which focuses on \textbf{detecting salient objects align with users' proactive needs when user have needs}. The main challenge for this new task is the lack of datasets for model training and testing.

cs.CV↗
arXiv2026

The chemRIXS Instrument for the LCLS-II X-Ray Free Electron Laser

The chemRIXS instrument at the Linac Coherent Light Source offers new opportunities for studying solution-phase systems with time-resolved soft X-ray spectroscopy through the recently commissioned high-repetition-rate LCLS-II X-ray free electron laser. The orders-of-magnitude X-ray flux improvement provided by the superconducting accelerator, combined with corresponding advances in the optical laser system and the liquid jet recirculation system, enables studies on dilute systems with high signal-to-noise compared to what was possible with the LCLS-I copper accelerator. These capabilities open up time-resolved X-ray absorption spectroscopy and resonant inelastic X-ray scattering to entirely new classes of samples, as well as enabling the development of new soft X-ray spectroscopies on liquid samples. An overview of the beamline components and the first LCLS-II commissioning results are presented.

physics.ins-det↗
arXiv2026

GDPO-SR: Group Direct Preference Optimization for One-Step Generative Image Super-Resolution

Recently, reinforcement learning (RL) has been employed for improving generative image super-resolution (ISR) performance. However, the current efforts are focused on multi-step generative ISR, while one-step generative ISR remains underexplored due to its limited stochasticity. In addition, RL methods such as Direct Preference Optimization (DPO) require the generation of positive and negative sample pairs offline, leading to a limited number of samples, while Group Relative Policy Optimization (GRPO) only calculates the likelihood of the entire image, ignoring local details that are crucial for ISR. In this paper, we propose Group Direct Preference Optimization (GDPO), a novel approach to integrate RL into one-step generative ISR model training. First, we introduce a noise-aware one-step diffusion model that can generate diverse ISR outputs. To prevent performance degradation caused by noise injection, we introduce an unequal-timestep strategy to decouple the timestep of noise addition from that of diffusion. We then present the GDPO strategy, which integrates the principle of GRPO into DPO, to calculate the group-relative advantage of each online generated sample for model optimization. Meanwhile, an attribute-aware reward function is designed to dynamically evaluate the score of each sample based on its statistics of smooth and texture areas. Experiments demonstrate the effectiveness of GDPO in enhancing the performance of one-step generative ISR models. Code: https://github.com/Joyies/GDPO.

cs.CV↗
arXiv2026

Acoustic Signatures of Pinch-Off Cavities During Water-Entry

This study experimentally, numerically, and theoretically investigates the cavity/bubble dynamics and radiated acoustics during the water entry of a centimeter-scale cylindrical projectile with a conical nose. Experiments were conducted in a laboratory tank, employing synchronized high-speed imaging and hydrophone measurements to characterize the cavity closure modes and their resultant acoustic signatures across a range of Froude numbers. The acoustic signal features a weak radiated signal upon impact, followed by significant pressure oscillations spanning more than 20 cycles in the flow field after cavity elongation and pinch-off. A numerical model based on the Finite Volume Method (FVM) successfully captures these physical processes. Subsequently, a semi-theoretical model that incorporates the projectile's boundary effect is developed from potential flow theory. The model not only yields a dominant cavity oscillation frequency that agrees well with experimental data, but also reveals that the boundary effect leads to a cavity oscillation frequency markedly higher than the Minnaert frequency of an equivalent-volume ellipsoidal bubble containing an internal rigid core. The dominant cavity frequency falls nearly linearly with Fr, governed by nose geometry and projectile inertia. This study clarifies the underlying physics connecting cavity dynamics during water entry to underwater acoustic radiation.

physics.flu-dyn↗
arXiv2026

Conditionally Whitened Generative Models for Probabilistic Time Series Forecasting

Probabilistic forecasting of multivariate time series is challenging due to non-stationarity, inter-variable dependencies, and distribution shifts. While recent diffusion and flow matching models have shown promise, they often ignore informative priors such as conditional means and covariances. In this work, we propose Conditionally Whitened Generative Models (CW-Gen), a framework that incorporates prior information through conditional whitening. Theoretically, we establish sufficient conditions under which replacing the traditional terminal distribution of diffusion models, namely the standard multivariate normal, with a multivariate normal distribution parameterized by estimators of the conditional mean and covariance improves sample quality. Guided by this analysis, we design a novel Joint Mean-Covariance Estimator (JMCE) that simultaneously learns the conditional mean and sliding-window covariance. Building on JMCE, we introduce Conditionally Whitened Diffusion Models (CW-Diff) and extend them to Conditionally Whitened Flow Matching (CW-Flow). Experiments on five real-world datasets with six state-of-the-art generative models demonstrate that CW-Gen consistently enhances predictive performance, capturing non-stationary dynamics and inter-variable correlations more effectively than prior-free approaches. Empirical results further demonstrate that CW-Gen can effectively mitigate the effects of distribution shift.

stat.ML↗
arXiv2026

Restoration Adaptation for Semantic Segmentation on Low Quality Images

In real-world scenarios, the performance of semantic segmentation often deteriorates when processing low-quality (LQ) images, which may lack clear semantic structures and high-frequency details. Although image restoration techniques offer a promising direction for enhancing degraded visual content, conventional real-world image restoration (Real-IR) models primarily focus on pixel-level fidelity and often fail to recover task-relevant semantic cues, limiting their effectiveness when directly applied to downstream vision tasks. Conversely, existing segmentation models trained on high-quality data lack robustness under real-world degradations. In this paper, we propose Restoration Adaptation for Semantic Segmentation (RASS), which effectively integrates semantic image restoration into the segmentation process, enabling high-quality semantic segmentation on the LQ images directly. Specifically, we first propose a Semantic-Constrained Restoration (SCR) model, which injects segmentation priors into the restoration model by aligning its cross-attention maps with segmentation masks, encouraging semantically faithful image reconstruction. Then, RASS transfers semantic restoration knowledge into segmentation through LoRA-based module merging and task-specific fine-tuning, thereby enhancing the model's robustness to LQ images. To validate the effectiveness of our framework, we construct a real-world LQ image segmentation dataset with high-quality annotations, and conduct extensive experiments on both synthetic and real-world LQ benchmarks. The results show that SCR and RASS significantly outperform state-of-the-art methods in segmentation and restoration tasks. Code, models, and datasets will be available at https://github.com/Ka1Guan/RASS.git.

cs.CV↗
arXiv2026

Fast Flow Matching based Conditional Independence Tests for Causal Discovery

Constraint-based causal discovery methods require a large number of conditional independence (CI) tests, which severely limits their practical applicability due to high computational complexity. Therefore, it is crucial to design an algorithm that accelerates each individual test. To this end, we propose the Flow Matching-based Conditional Independence Test (FMCIT). The proposed test leverages the high computational efficiency of flow matching and requires the model to be trained only once throughout the entire causal discovery procedure, substantially accelerating causal discovery. According to numerical experiments, FMCIT effectively controls type-I error and maintains high testing power under the alternative hypothesis, even in the presence of high-dimensional conditioning sets. In addition, we further integrate FMCIT into a two-stage guided PC skeleton learning framework, termed GPC-FMCIT, which combines fast screening with guided, budgeted refinement using FMCIT. This design yields explicit bounds on the number of CI queries while maintaining high statistical power. Experiments on synthetic and real-world causal discovery tasks demonstrate favorable accuracy-efficiency trade-offs over existing CI testing methods and PC variants.

cs.LG↗
arXiv2026

Research on a Camera Position Measurement Method based on a Parallel Perspective Error Transfer Model

Camera pose estimation from sparse correspondences is a fundamental problem in geometric computer vision and remains particularly challenging in near-field scenarios, where strong perspective effects and heterogeneous measurement noise can significantly degrade the stability of analytic PnP solutions. In this paper, we present a geometric error propagation framework for camera pose estimation based on a parallel perspective approximation. By explicitly modeling how image measurement errors propagate through perspective geometry, we derive an error transfer model that characterizes the relationship between feature point distribution, camera depth, and pose estimation uncertainty. Building on this analysis, we develop a pose estimation method that leverages parallel perspective initialization and error-aware weighting within a Gauss-Newton optimization scheme, leading to improved robustness in proximity operations. Extensive experiments on both synthetic data and real-world images, covering diverse conditions such as strong illumination, surgical lighting, and underwater low-light environments, demonstrate that the proposed approach achieves accuracy and robustness comparable to state-of-the-art analytic and iterative PnP methods, while maintaining high computational efficiency. These results highlight the importance of explicit geometric error modeling for reliable camera pose estimation in challenging near-field settings.

cs.RO↗
arXiv2026

LLM4Fluid: Large Language Models as Generalizable Neural Solvers for Fluid Dynamics

Deep learning has emerged as a promising paradigm for spatio-temporal modeling of fluid dynamics. However, existing approaches often suffer from limited generalization to unseen flow conditions and typically require retraining when applied to new scenarios. In this paper, we present LLM4Fluid, a spatio-temporal prediction framework that leverages Large Language Models (LLMs) as generalizable neural solvers for fluid dynamics. The framework first compresses high-dimensional flow fields into a compact latent space via reduced-order modeling enhanced with a physics-informed disentanglement mechanism, effectively mitigating spatial feature entanglement while preserving essential flow structures. A pretrained LLM then serves as a temporal processor, autoregressively predicting the dynamics of physical sequences with time series prompts. To bridge the modality gap between prompts and physical sequences, which can otherwise degrade prediction accuracy, we propose a dedicated modality alignment strategy that resolves representational mismatch and stabilizes long-term prediction. Extensive experiments across diverse flow scenarios demonstrate that LLM4Fluid functions as a robust and generalizable neural solver without retraining, achieving state-of-the-art accuracy while exhibiting powerful zero-shot and in-context learning capabilities. Code and datasets are publicly available at https://github.com/qisongxiao/LLM4Fluid.

cs.LG↗
arXiv2026

Interpretability and Individuality in Knee MRI: Patient-Specific Radiomic Fingerprint with Reconstructed Healthy Personas

For automated assessment of knee MRI scans, both accuracy and interpretability are essential for clinical use and adoption. Traditional radiomics rely on predefined features chosen at the population level; while more interpretable, they are often too restrictive to capture patient-specific variability and can underperform end-to-end deep learning (DL). To address this, we propose two complementary strategies that bring individuality and interpretability: radiomic fingerprints and healthy personas. First, a radiomic fingerprint is a dynamically constructed, patient-specific feature set derived from MRI. Instead of applying a uniform population-level signature, our model predicts feature relevance from a pool of candidate features and selects only those most predictive for each patient, while maintaining feature-level interpretability. This fingerprint can be viewed as a latent-variable model of feature usage, where an image-conditioned predictor estimates usage probabilities and a transparent logistic regression with global coefficients performs classification. Second, a healthy persona synthesises a pathology-free baseline for each patient using a diffusion model trained to reconstruct healthy knee MRIs. Comparing features extracted from pathological images against their personas highlights deviations from normal anatomy, enabling intuitive, case-specific explanations of disease manifestations. We systematically compare fingerprints, personas, and their combination across three clinical tasks. Experimental results show that both approaches yield performance comparable to or surpassing state-of-the-art DL models, while supporting interpretability at multiple levels. Case studies further illustrate how these perspectives facilitate human-explainable biomarker discovery and pathology localisation.

cs.CV↗
arXiv2026

Supersonic jet dynamics from two-cavitation-bubble interactions: acceleration, tip fragmentation and penetration

This study experimentally and numerically investigates the dynamics of a high-speed liquid jet generated from the interaction of two tandem cavitation bubbles, termed bubble 1 and bubble 2, depending on their generation sequence. In our experiments, two near-identical, highly-energized cavitation bubbles were generated using an underwater electric discharge method, and their transient interactions were captured using a high-speed camera. We identify three distinct jet regimes that emerge from the tip of bubble 2: conical, umbrella-shaped, and spraying jets, characterized by variations in the initial bubble-bubble distance and the initiation time difference. Our numerical simulations using both Volume of Fluid and Boundary Integral methods reproduce the experimental observations quite well and explain the mechanism of jet acceleration. We show that the transition between the regimes is governed by the spatiotemporal characteristics of the pressure wave induced by the collapse of bubble 1, which impacts the high-curvature tip of bubble 2. Specifically, a conical jet forms when the pressure wave impacts the bubble tip prior to its contraction, while an umbrella-shaped jet develops when this impact occurs after the contraction. The spraying jets result from the breakup of the bubble tip, exhibiting mist-like and needle-like morphologies with velocities ranging from 10 to over 1200 m/s. Remarkably, we observe that the penetration distance of spraying jets exceeds ten times the maximum bubble radius, making them ideal for long-range, controlled fluid delivery. Finally, phase diagrams for jet velocity and penetration distance in the $γ-θ$ parameter space are established to provide a practical reference for biomedical applications, such as needle-free injection and micro-pumping.

physics.flu-dyn↗
arXiv2026

Exploration of Altermagnetism in $\mathrm{RuO_{2}}$

The fundamental role of magnetic materials in modern science and technology has driven a rapid surge in research on unconventional magnetism in recent years. In particular, altermagnets, which simultaneously exhibit zero net magnetization in real space and anisotropic spin splitting in momentum space, have garnered significant interest for both fundamental physics and technological applications. Among these, $\mathrm{RuO_{2}}$ stands as the pioneering and most extensively studied altermagnet. While the intrinsic magnetic order of $\mathrm{RuO_{2}}$ is still a subject of active debate, numerous exotic phenomena characteristic of altermagnetism have been observed in $\mathrm{RuO_{2}}$ samples. In this review, we explore each facet of the altermagnetism through specific case studies in $\mathrm{RuO_{2}}$, systematically surveying its crystal and magnetic structures, electronic band properties, and transport phenomena. We critically assess the debate surrounding the intrinsic magnetism in $\mathrm{RuO_{2}}$, incorporating evidence from altermagnetic signatures in transport, as well as contrasting results from magnetic and spectroscopic measurements. Finally, possible future research directions in this field are discussed.

cond-mat.mes-hall↗
arXiv2026

Multi-Subspace Multi-Modal Modeling for Diffusion Models: Estimation, Convergence and Mixture of Experts

Recently, diffusion models have achieved a great performance with a small dataset of size $n$ and a fast optimization process. However, the estimation error of diffusion models suffers from the curse of dimensionality $n^{-1/D}$ with the data dimension $D$. Since images are usually a union of low-dimensional manifolds, current works model the data as a union of linear subspaces with Gaussian latent and achieve a $1/\sqrt{n}$ bound. Though this modeling reflects the multi-manifold property, the Gaussian latent can not capture the multi-modal property of the latent manifold. To bridge this gap, we propose the mixture subspace of low-rank mixture of Gaussian (MoLR-MoG) modeling, which models the target data as a union of $K$ linear subspaces, and each subspace admits a mixture of Gaussian latent ($n_k$ modals with dimension $d_k$). With this modeling, the corresponding score function naturally has a mixture of expert (MoE) structure, captures the multi-modal information, and contains nonlinear property. We first conduct real-world experiments to show that the generation results of MoE-latent MoG NN are much better than MoE-latent Gaussian score. Furthermore, MoE-latent MoG NN achieves a comparable performance with MoE-latent Unet with $10 \times$ parameters. These results indicate that the MoLR-MoG modeling is reasonable and suitable for real-world data. After that, based on such MoE-latent MoG score, we provide a $R^4\sqrt{Σ_{k=1}^Kn_k}\sqrt{Σ_{k=1}^Kn_kd_k}/\sqrt{n}$ estimation error, which escapes the curse of dimensionality by using data structure. Finally, we study the optimization process and prove the convergence guarantee under the MoLR-MoG modeling. Combined with these results, under a setting close to real-world data, this work explains why diffusion models only require a small training sample and enjoy a fast optimization process to achieve a great performance.

cs.LG↗
arXiv2026

GenCAMO: Scene-Graph Contextual Decoupling for Environment-aware and Mask-free Camouflage Image-Dense Annotation Generation

Conceal dense prediction (CDP), especially RGB-D camouflage object detection and open-vocabulary camouflage object segmentation, plays a crucial role in advancing the understanding and reasoning of complex camouflage scenes. However, high-quality and large-scale camouflage datasets with dense annotation remain scarce due to expensive data collection and labeling costs. To address this challenge, we explore leveraging generative models to synthesize realistic camouflage image-dense data for training CDP models with fine-grained representations, prior knowledge, and auxiliary reasoning. Concretely, our contributions are threefold: (i) we introduce GenCAMO-DB, a large-scale camouflage dataset with multi-modal annotations, including depth maps, scene graphs, attribute descriptions, and text prompts; (ii) we present GenCAMO, an environment-aware and mask-free generative framework that produces high-fidelity camouflage image-dense annotations; (iii) extensive experiments across multiple modalities demonstrate that GenCAMO significantly improves dense prediction performance on complex camouflage scenes by providing high-quality synthetic data. The code and datasets will be released after paper acceptance.

cs.CV↗
arXiv2026

Linear magnetoresistance of two-dimensional massless Dirac fermions in the quantum limit

Linear magnetoresistance is a hallmark of 3D Weyl metals in the quantum limit. Recently, a pronounced linear magnetoresistance has also been observed in 2D graphene [Xin et al., Nature 616, 270 (2023)]. However, a comprehensive theoretical understanding remains elusive. By employing the self-consistent Born approximation, we derive the analytical expressions for the magnetoresistivity of 2D massless Dirac fermions in the quantum limit. Notably, our result recovers the minimum conductivity in the clean limit and reveals a linear dependence of resistivity on the magnetic field for Gaussian impurity potentials, in quantitative agreement with experiments. These findings shed light on the magnetoresistance behavior of 2D Dirac fermions under ultra-high magnetic fields.

cond-mat.mes-hall↗