arXiv subjects

Tianyu Zhang

Publications and source records attributed to Tianyu Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2025

An expanded evaluation matrix for the entropy-weight TODIM method to reduce the rank reversal probability and its application in selecting energy storage technology

The TODIM method (an acronym in Portuguese for interactive and multiple criteria decision-making) with entropy weights is influenced by rank reversal, a phenomenon where the order of two alternatives changes following the addition of another alternative. Research on rank reversal has predominantly focused on single decision-making methods. To the best of our knowledge, the reduction of rank reversal probability in hybrid methods, such as the entropy-weight TODIM method, remains an unresolved challenge. To address this, this paper introduces the expanded evaluation matrix, which incorporates virtual alternatives, to reduce the probability of rank reversal in the entropy-weight TODIM method. A simulation study is conducted to assess the effectiveness of the expanded evaluation matrix in mitigating rank reversal. The results demonstrate that the expanded evaluation matrix significantly reduces the rank reversal probability. A case study on selecting energy storage technology showcases the potential real-world applications of the expanded evaluation matrix. The reliability of the expanded evaluation matrix is further validated through sensitivity and comparative analyses. Given the simplicity and ease of implementation of the expanded evaluation matrix, it can be readily adapted to other decision-making methods and holds substantial potential for broad application.

math.OC↗

arXiv2025

Combined-distance-based score function of cognitive fuzzy sets and its application in lung cancer pain evaluation

In decision making, the cognitive fuzzy set (CFS) is a useful tool in expressing experts' complex assessments of alternatives. The distance of CFS, which plays an important role in decision analyses, is necessary when the CFS is applied in solving practical issues. However, as far as we know, the studies on the distance of CFS are few, and the current Minkowski distance of CFS ignores the hesitancy degree of CFS, which might cause errors. To fill the gap of the studies on the distance of CFS, because of the practicality of the Hausdorff distance, this paper proposes the improved cognitive fuzzy Minkowski (CF-IM) distance and the cognitive fuzzy Hausdorff (CF-H) distance to enrich the studies on the distance of CFS. It is found that the anti-perturbation ability of the CF-H distance is stronger than that of the CF-IM distance, but the information utilization of the CF-IM distance is higher than that of the CF-H distance. To balance the anti-perturbation ability and information utilization of the CF-IM distance and CF-H distance, the cognitive fuzzy combined (CF-C) distance is proposed by establishing the linear combination of the CF-IM distance and CF-H distance. Based on the CF-C distance, a combined-distanced-based score function of CFS is proposed to compare CFSs. The proposed score function is employed in lung cancer pain evaluation issues. The sensitivity and comparison analyses demonstrate the reliability and advantages of the proposed methods.

math.OC↗

arXiv2025

Learning User Preferences for Image Generation Model

User preference prediction requires a comprehensive and accurate understanding of individual tastes. This includes both surface-level attributes, such as color and style, and deeper content-related aspects, such as themes and composition. However, existing methods typically rely on general human preferences or assume static user profiles, often neglecting individual variability and the dynamic, multifaceted nature of personal taste. To address these limitations, we propose an approach built upon Multimodal Large Language Models, introducing contrastive preference loss and preference tokens to learn personalized user preferences from historical interactions. The contrastive preference loss is designed to effectively distinguish between user ''likes'' and ''dislikes'', while the learnable preference tokens capture shared interest representations among existing users, enabling the model to activate group-specific preferences and enhance consistency across similar users. Extensive experiments demonstrate our model outperforms other methods in preference prediction accuracy, effectively identifying users with similar aesthetic inclinations and providing more precise guidance for generating images that align with individual tastes. The project page is \texttt{https://learn-user-pref.github.io/}.

cs.CV↗

arXiv2025

Adaptique: Multi-objective and Context-aware Online Adaptation of Selection Techniques in Virtual Reality

Selection is a fundamental task that is challenging in virtual reality due to issues such as distant and small targets, occlusion, and target-dense environments. Previous research has tackled these challenges through various selection techniques, but complicates selection and can be seen as tedious outside of their designed use case. We present Adaptique, an adaptive model that infers and switches to the most optimal selection technique based on user and environmental information. Adaptique considers contextual information such as target size, distance, occlusion, and user posture combined with four objectives: speed, accuracy, comfort, and familiarity which are based on fundamental predictive models of human movement for technique selection. This enables Adaptique to select simple techniques when they are sufficiently efficient and more advanced techniques when necessary. We show that Adaptique is more preferred and performant than single techniques in a user study, and demonstrate Adaptique's versatility in an application.

cs.HC↗

arXiv2025

Aligning Constraint Generation with Design Intent in Parametric CAD

We adapt alignment techniques from reasoning LLMs to the task of generating engineering sketch constraints found in computer-aided design (CAD) models. Engineering sketches consist of geometric primitives (e.g. points, lines) connected by constraints (e.g. perpendicular, tangent) that define the relationships between them. For a design to be easily editable, the constraints must effectively capture design intent, ensuring the geometry updates predictably when parameters change. Although current approaches can generate CAD designs, an open challenge remains to align model outputs with design intent, we label this problem 'design alignment'. A critical first step towards aligning generative CAD models is to generate constraints which fully-constrain all geometric primitives, without over-constraining or distorting sketch geometry. Using alignment techniques to train an existing constraint generation model with feedback from a constraint solver, we are able to fully-constrain 93% of sketches compared to 34% when using a naive supervised fine-tuning (SFT) baseline and only 8.9% without SFT. Our approach can be applied to any existing constraint generation model and sets the stage for further research bridging alignment strategies between the language and design domains. Additional results can be found at https://autodeskailab.github.io/aligning-constraint-generation/.

cs.LG↗

arXiv2025

Enhancing Reward Models for High-quality Image Generation: Beyond Text-Image Alignment

Contemporary image generation systems have achieved high fidelity and superior aesthetic quality beyond basic text-image alignment. However, existing evaluation frameworks have failed to evolve in parallel. This study reveals that human preference reward models fine-tuned based on CLIP and BLIP architectures have inherent flaws: they inappropriately assign low scores to images with rich details and high aesthetic value, creating a significant discrepancy with actual human aesthetic preferences. To address this issue, we design a novel evaluation score, ICT (Image-Contained-Text) score, that achieves and surpasses the objectives of text-image alignment by assessing the degree to which images represent textual content. Building upon this foundation, we further train an HP (High-Preference) score model using solely the image modality to enhance image aesthetics and detail quality while maintaining text-image alignment. Experiments demonstrate that the proposed evaluation model improves scoring accuracy by over 10\% compared to existing methods, and achieves significant results in optimizing state-of-the-art text-to-image models. This research provides theoretical and empirical support for evolving image generation technology toward higher-order human aesthetic preferences. Code is available at https://github.com/BarretBa/ICTHP.

cs.CV↗

arXiv2025

Interactive Drawing Guidance for Anime Illustrations with Diffusion Model

Creating high-quality anime illustrations presents notable challenges, particularly for beginners, due to the intricate styles and fine details inherent in anime art. We present an interactive drawing guidance system specifically designed for anime illustrations to address this issue. It offers real-time guidance to help users refine their work and streamline the creative process. Our system is built upon the StreamDiffusion pipeline to deliver real-time drawing assistance. We fine-tune Stable Diffusion with LoRA to synthesize anime style RGB images from user-provided hand-drawn sketches and prompts. Leveraging the Informative Drawings model, we transform these RGB images into rough sketches, which are further refined into structured guidance sketches using a custom-designed optimizer. The proposed system offers precise, real-time guidance aligned with the creative intent of the user, significantly enhancing both the efficiency and accuracy of the drawing process. To assess the effectiveness of our approach, we conducted a user study, gathering empirical feedback on both system performance and interface usability.

cs.GR↗

arXiv2025

Modeling the impacts of galaxy intrinsic alignments on weak lensing peak statistics

Weak gravitational lensing (WL) peak statistics capture cosmic non-linear structures and can provide additional cosmological information complementary to cosmic shear two-point correlation analyses. They have been applied to different WL surveys successfully. To further facilitate their high precision applications, it is very timely to investigate the impacts of different systematics on WL peak statistics and how to mitigate them. Concerning the influence from galaxy intrinsic alignments (IAs), in this paper, we develop a theoretical model for WL high peaks taking into account the IA effects. It is an extension of our previous halo-based model. The IA corrections mainly include the modification of the lensing profile of clusters of galaxies due to the alignments of satellite galaxies and the additional shape noise correlations. We validate our model using simulations with the semi-analytical galaxy formation. We consider the cases where the satellite galaxies are averagely radially aligned toward the centers of their host clusters but with different dispersions $σ_θ$. We show that our model works well for $σ_θ>45^{\circ}$. If the IA corrections are not included in the model, for the Euclid/CSST-like source galaxy distribution and the survey area of $\sim 1000 deg^2$, the IA induced bias on $S_8$ can reach $\sim 8σ$ even for $σ_θ=75^{\circ}$. With our model, not only the bias can be well mitigated, but also we can constrain the satellite IA to the level of {\bf $σ(σ_θ)\sim \pm 24^{\circ}$} simultaneously from WL high peak analyses alone using data from such a survey.

astro-ph.CO↗

arXiv2025

DpDNet: An Dual-Prompt-Driven Network for Universal PET-CT Segmentation

PET-CT lesion segmentation is challenging due to noise sensitivity, small and variable lesion morphology, and interference from physiological high-metabolic signals. Current mainstream approaches follow the practice of one network solving the segmentation of multiple cancer lesions by treating all cancers as a single task. However, this overlooks the unique characteristics of different cancer types. Considering the specificity and similarity of different cancers in terms of metastatic patterns, organ preferences, and FDG uptake intensity, we propose DpDNet, a Dual-Prompt-Driven network that incorporates specific prompts to capture cancer-specific features and common prompts to retain shared knowledge. Additionally, to mitigate information forgetting caused by the early introduction of prompts, prompt-aware heads are employed after the decoder to adaptively handle multiple segmentation tasks. Experiments on a PET-CT dataset with four cancer types show that DpDNet outperforms state-of-the-art models. Finally, based on the segmentation results, we calculated MTV, TLG, and SUVmax for breast cancer survival analysis. The results suggest that DpDNet has the potential to serve as a valuable tool for personalized risk stratification, supporting clinicians in optimizing treatment strategies and improving outcomes. Code is available at https://github.com/XinglongLiang08/DpDNet.

eess.IV↗

arXiv2025

Online Estimation with Rolling Validation: Adaptive Nonparametric Estimation with Streaming Data

Online nonparametric estimators are gaining popularity due to their efficient computation and competitive generalization abilities. An important example includes variants of stochastic gradient descent. These algorithms often take one sample point at a time and incrementally update the parameter estimate of interest. In this work, we consider model selection/hyperparameter tuning for such online algorithms. We propose a weighted rolling validation procedure, an online variant of leave-one-out cross-validation, that costs minimal extra computation for many typical stochastic gradient descent estimators and maintains their online nature. Similar to batch cross-validation, it can boost base estimators to achieve better heuristic performance and adaptive convergence rate. Our analysis is straightforward, relying mainly on some general statistical stability assumptions. The simulation study underscores the significance of diverging weights in practice and demonstrates its favorable sensitivity even when there is only a slim difference between candidate estimators.

math.ST↗

arXiv2025

Latent Posterior-Mean Rectified Flow for Higher-Fidelity Perceptual Face Restoration

The Perception-Distortion tradeoff (PD-tradeoff) theory suggests that face restoration algorithms must balance perceptual quality and fidelity. To achieve minimal distortion while maintaining perfect perceptual quality, Posterior-Mean Rectified Flow (PMRF) proposes a flow based approach where source distribution is minimum distortion estimations. Although PMRF is shown to be effective, its pixel-space modeling approach limits its ability to align with human perception, where human perception is defined as how humans distinguish between two image distributions. In this work, we propose Latent-PMRF, which reformulates PMRF in the latent space of a variational autoencoder (VAE), facilitating better alignment with human perception during optimization. By defining the source distribution on latent representations of minimum distortion estimation, we bound the minimum distortion by the VAE's reconstruction error. Moreover, we reveal the design of VAE is crucial, and our proposed VAE significantly outperforms existing VAEs in both reconstruction and restoration. Extensive experiments on blind face restoration demonstrate the superiority of Latent-PMRF, offering an improved PD-tradeoff compared to existing methods, along with remarkable convergence efficiency, achieving a 5.79X speedup over PMRF in terms of FID. Our code will be available as open-source.

cs.CV↗

arXiv2025

StableCodec: Taming One-Step Diffusion for Extreme Image Compression

Diffusion-based image compression has shown remarkable potential for achieving ultra-low bitrate coding (less than 0.05 bits per pixel) with high realism, by leveraging the generative priors of large pre-trained text-to-image diffusion models. However, current approaches require a large number of denoising steps at the decoder to generate realistic results under extreme bitrate constraints, limiting their application in real-time compression scenarios. Additionally, these methods often sacrifice reconstruction fidelity, as diffusion models typically fail to guarantee pixel-level consistency. To address these challenges, we introduce StableCodec, which enables one-step diffusion for high-fidelity and high-realism extreme image compression with improved coding efficiency. To achieve ultra-low bitrates, we first develop an efficient Deep Compression Latent Codec to transmit a noisy latent representation for a single-step denoising process. We then propose a Dual-Branch Coding Structure, consisting of a pair of auxiliary encoder and decoder, to enhance reconstruction fidelity. Furthermore, we adopt end-to-end optimization with joint bitrate and pixel-level constraints. Extensive experiments on the CLIC 2020, DIV2K, and Kodak dataset demonstrate that StableCodec outperforms existing methods in terms of FID, KID and DISTS by a significant margin, even at bitrates as low as 0.005 bits per pixel, while maintaining strong fidelity. Additionally, StableCodec achieves inference speeds comparable to mainstream transform coding schemes. All source code are available at https://github.com/LuizScarlet/StableCodec.

eess.IV↗

arXiv2025

Supporting Deterministic Traffic on Standard NICs

Networked mission-critical applications (e.g., avionic control and industrial automation systems) require deterministic packet transmissions to support a range of sensing and control tasks with stringent timing constraints. While specialized network infrastructure (e.g., time-sensitive networking (TSN) switches) provides deterministic data transport across the network, achieving strict end-to-end timing guarantees requires equally capable end devices to support deterministic traffic. These end devices, however, often employ general-purpose computing platforms like standard PCs, which lack native support for deterministic traffic and suffer from unpredictable delays introduced by their software stack and system architecture. Although specialized NICs with hardware scheduling offload can mitigate this problem, the limited compatibility hinders their widespread adoption, particularly for cost-sensitive applications or in legacy devices. To fill this gap, this paper proposes a novel software-based driver model, namely KeepON, to enable the support of deterministic packet transmissions on end devices equipped with standard NICs. The key idea of KeepON is to have the NIC keep on transmitting fixed-size data chunks as placeholders, thereby maintaining a predictable temporal transmission pattern. The real-time packets generated by the mission-critical application(s) will then be precisely inserted into this stream by replacing placeholders at the designated position to ensure their accurate transmission time. We implement and evaluate KeepON by modifying the network driver on a Raspberry Pi using its standard NIC. Our experiments demonstrate that KeepON can achieve x162 times scheduling accuracy comparable to its default driver, and x2.6 times compared to hardware-based solution, thus enabling precise timing control on standard commodity hardware.

cs.NI↗

arXiv2025

Adaptive Projected Two-Sample Comparisons for Single-Cell Gene Expression Data

We study high-dimensional two-sample mean comparison and address the curse of dimensionality through data-adaptive projections. Leveraging the low-dimensional and localized signal structures commonly seen in single-cell genomics data, our first proposed method identifies a sparse, informative low-dimensional subspace and then performs statistical inference restricted to this subspace. To address the double-dipping issue -- arising from using the same data for projection and inference -- we develop a debiased projected estimator using the semiparametric double-machine learning framework. The resulting inference not only has the usual frequentist validity but also provides useful information on the potential location of the signal due to the sparsity of the projection. Our second method uses a more flexible projection scheme to improve the power against the global null hypothesis and avoid the degeneracy issue commonly faced by existing methods. It is particularly useful when debiasing is practically challenging or when the informative signal is not well-captured by the subspace. Experiments on synthetic data and real datasets demonstrate the theoretical promise and interpretability of the proposed methods.

stat.ME↗

arXiv2025

NTIRE 2025 Challenge on RAW Image Restoration and Super-Resolution

This paper reviews the NTIRE 2025 RAW Image Restoration and Super-Resolution Challenge, highlighting the proposed solutions and results. New methods for RAW Restoration and Super-Resolution could be essential in modern Image Signal Processing (ISP) pipelines, however, this problem is not as explored as in the RGB domain. The goal of this challenge is two fold, (i) restore RAW images with blur and noise degradations, (ii) upscale RAW Bayer images by 2x, considering unknown noise and blur. In the challenge, a total of 230 participants registered, and 45 submitted results during thee challenge period. This report presents the current state-of-the-art in RAW Restoration.

eess.IV↗

arXiv2025

FairZK: A Scalable System to Prove Machine Learning Fairness in Zero-Knowledge

With the rise of machine learning techniques, ensuring the fairness of decisions made by machine learning algorithms has become of great importance in critical applications. However, measuring fairness often requires full access to the model parameters, which compromises the confidentiality of the models. In this paper, we propose a solution using zero-knowledge proofs, which allows the model owner to convince the public that a machine learning model is fair while preserving the secrecy of the model. To circumvent the efficiency barrier of naively proving machine learning inferences in zero-knowledge, our key innovation is a new approach to measure fairness only with model parameters and some aggregated information of the input, but not on any specific dataset. To achieve this goal, we derive new bounds for the fairness of logistic regression and deep neural network models that are tighter and better reflecting the fairness compared to prior work. Moreover, we develop efficient zero-knowledge proof protocols for common computations involved in measuring fairness, including the spectral norm of matrices, maximum, absolute value, and fixed-point arithmetic. We have fully implemented our system, FairZK, that proves machine learning fairness in zero-knowledge. Experimental results show that FairZK is significantly faster than the naive approach and an existing scheme that use zero-knowledge inferences as a subroutine. The prover time is improved by 3.1x--1789x depending on the size of the model and the dataset. FairZK can scale to a large model with 47 million parameters for the first time, and generates a proof for its fairness in 343 seconds. This is estimated to be 4 orders of magnitude faster than existing schemes, which only scale to small models with hundreds to thousands of parameters.

cs.LG↗

arXiv2025

PHYBench: Holistic Evaluation of Physical Perception and Reasoning in Large Language Models

Current benchmarks for evaluating the reasoning capabilities of Large Language Models (LLMs) face significant limitations: task oversimplification, data contamination, and flawed evaluation items. These deficiencies necessitate more rigorous assessment methods. To address these limitations, we introduce PHYBench, a benchmark of 500 original physics problems ranging from high school to Physics Olympiad difficulty. PHYBench addresses data contamination through original content and employs a systematic curation pipeline to eliminate flawed items. Evaluations show that PHYBench activates more tokens and provides stronger differentiation between reasoning models compared to other baselines like AIME 2024, OlympiadBench and GPQA. Even the best-performing model, Gemini 2.5 Pro, achieves only 36.9% accuracy compared to human experts' 61.9%. To further enhance evaluation precision, we introduce the Expression Edit Distance (EED) Score for mathematical expression assessment, which improves sample efficiency by 204% over binary scoring. Moreover, PHYBench effectively elicits multi-step and multi-condition reasoning, providing a platform for examining models' reasoning robustness, preferences, and deficiencies. The benchmark results and dataset are publicly available at https://www.phybench.cn/.

cs.CL↗

arXiv2025

Neural Co-Optimization of Structural Topology, Manufacturable Layers, and Path Orientations for Fiber-Reinforced Composites

We propose a neural network-based computational framework for the simultaneous optimization of structural topology, curved layers, and path orientations to achieve strong anisotropic strength in fiber-reinforced thermoplastic composites while ensuring manufacturability. Our framework employs three implicit neural fields to represent geometric shape, layer sequence, and fiber orientation. This enables the direct formulation of both design and manufacturability objectives - such as anisotropic strength, structural volume, machine motion control, layer curvature, and layer thickness - into an integrated and differentiable optimization process. By incorporating these objectives as loss functions, the framework ensures that the resultant composites exhibit optimized mechanical strength while remaining its manufacturability for filament-based multi-axis 3D printing across diverse hardware platforms. Physical experiments demonstrate that the composites generated by our co-optimization method can achieve an improvement of up to 33.1% in failure loads compared to composites with sequentially optimized structures and manufacturing sequences.

cs.LG↗