arXiv subjects

Yuxuan Zhang

Publications and source records attributed to Yuxuan Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2025

NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement: Methods and Results

This paper presents a review for the NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement. The challenge comprises two tracks: (i) Efficient Video Quality Assessment (KVQ), and (ii) Diffusion-based Image Super-Resolution (KwaiSR). Track 1 aims to advance the development of lightweight and efficient video quality assessment (VQA) models, with an emphasis on eliminating reliance on model ensembles, redundant weights, and other computationally expensive components in the previous IQA/VQA competitions. Track 2 introduces a new short-form UGC dataset tailored for single image super-resolution, i.e., the KwaiSR dataset. It consists of 1,800 synthetically generated S-UGC image pairs and 1,900 real-world S-UGC images, which are split into training, validation, and test sets using a ratio of 8:1:1. The primary objective of the challenge is to drive research that benefits the user experience of short-form UGC platforms such as Kwai and TikTok. This challenge attracted 266 participants and received 18 valid final submissions with corresponding fact sheets, significantly contributing to the progress of short-form UGC VQA and image superresolution. The project is publicly available at https://github.com/lixinustc/KVQE- ChallengeCVPR-NTIRE2025.

eess.IV↗

arXiv2025

Distillation-Supervised Convolutional Low-Rank Adaptation for Efficient Image Super-Resolution

Convolutional neural networks (CNNs) have been widely used in efficient image super-resolution. However, for CNN-based methods, performance gains often require deeper networks and larger feature maps, which increase complexity and inference costs. Inspired by LoRA's success in fine-tuning large language models, we explore its application to lightweight models and propose Distillation-Supervised Convolutional Low-Rank Adaptation (DSCLoRA), which improves model performance without increasing architectural complexity or inference costs. Specifically, we integrate ConvLoRA into the efficient SR network SPAN by replacing the SPAB module with the proposed SConvLB module and incorporating ConvLoRA layers into both the pixel shuffle block and its preceding convolutional layer. DSCLoRA leverages low-rank decomposition for parameter updates and employs a spatial feature affinity-based knowledge distillation strategy to transfer second-order statistical information from teacher models (pre-trained SPAN) to student models (ours). This method preserves the core knowledge of lightweight models and facilitates optimal solution discovery under certain conditions. Experiments on benchmark datasets show that DSCLoRA improves PSNR and SSIM over SPAN while maintaining its efficiency and competitive image quality. Notably, DSCLoRA ranked first in the Overall Performance Track of the NTIRE 2025 Efficient Super-Resolution Challenge. Our code and models are made publicly available at https://github.com/Yaozzz666/DSCF-SR.

cs.CV↗

arXiv2025

Structuring Scientific Innovation: A Framework for Modeling and Discovering Impactful Knowledge Combinations

The emergence of large language models offers new possibilities for structured exploration of scientific knowledge. Rather than viewing scientific discovery as isolated ideas or content, we propose a structured approach that emphasizes the role of method combinations in shaping disruptive insights. Specifically, we investigate how knowledge unit--especially those tied to methodological design--can be modeled and recombined to yield research breakthroughs. Our proposed framework addresses two key challenges. First, we introduce a contrastive learning-based mechanism to identify distinguishing features of historically disruptive method combinations within problem-driven contexts. Second, we propose a reasoning-guided Monte Carlo search algorithm that leverages the chain-of-thought capability of LLMs to identify promising knowledge recombinations for new problem statements.Empirical studies across multiple domains show that the framework is capable of modeling the structural dynamics of innovation and successfully highlights combinations with high disruptive potential. This research provides a new path for computationally guided scientific ideation grounded in structured reasoning and historical data modeling.

cs.AI↗

arXiv2025

Enhanced Semantic Extraction and Guidance for UGC Image Super Resolution

Due to the disparity between real-world degradations in user-generated content(UGC) images and synthetic degradations, traditional super-resolution methods struggle to generalize effectively, necessitating a more robust approach to model real-world distortions. In this paper, we propose a novel approach to UGC image super-resolution by integrating semantic guidance into a diffusion framework. Our method addresses the inconsistency between degradations in wild and synthetic datasets by separately simulating the degradation processes on the LSDIR dataset and combining them with the official paired training set. Furthermore, we enhance degradation removal and detail generation by incorporating a pretrained semantic extraction model (SAM2) and fine-tuning key hyperparameters for improved perceptual fidelity. Extensive experiments demonstrate the superiority of our approach against state-of-the-art methods. Additionally, the proposed model won second place in the CVPR NTIRE 2025 Short-form UGC Image Super-Resolution Challenge, further validating its effectiveness. The code is available at https://github.c10pom/Moonsofang/NTIRE-2025-SRlab.

cs.CV↗

arXiv2025

Can Competition Enhance the Proficiency of Agents Powered by Large Language Models in the Realm of News-driven Time Series Forecasting?

Multi-agents-based news-driven time series forecasting is considered as a potential paradigm shift in the era of large language models (LLMs). The challenge of this task lies in measuring the influences of different news events towards the fluctuations of time series. This requires agents to possess stronger abilities of innovative thinking and the identifying misleading logic. However, the existing multi-agent discussion framework has limited enhancement on time series prediction in terms of optimizing these two capabilities. Inspired by the role of competition in fostering innovation, this study embeds a competition mechanism within the multi-agent discussion to enhance agents' capability of generating innovative thoughts. Furthermore, to bolster the model's proficiency in identifying misleading information, we incorporate a fine-tuned small-scale LLM model within the reflective stage, offering auxiliary decision-making support. Experimental results confirm that the competition can boost agents' capacity for innovative thinking, which can significantly improve the performances of time series prediction. Similar to the findings of social science, the intensity of competition within this framework can influence the performances of agents, providing a new perspective for studying LLMs-based multi-agent systems.

cs.AI↗

arXiv2025

Beyond Words: AuralLLM and SignMST-C for Sign Language Production and Bidirectional Accessibility

Sign language is the primary communication mode for 72 million hearing-impaired individuals worldwide, necessitating effective bidirectional Sign Language Production and Sign Language Translation systems. However, functional bidirectional systems require a unified linguistic environment, hindered by the lack of suitable unified datasets, particularly those providing the necessary pose information for accurate Sign Language Production (SLP) evaluation. Concurrently, current SLP evaluation methods like back-translation ignore pose accuracy, and high-quality coordinated generation remains challenging. To create this crucial environment and overcome these challenges, we introduce CNText2Sign and CNSign, which together constitute the first unified dataset aimed at supporting bidirectional accessibility systems for Chinese sign language; CNText2Sign provides 15,000 natural language-to-sign mappings and standardized skeletal keypoints for 8,643 vocabulary items supporting pose assessment. Building upon this foundation, we propose the AuraLLM model, which leverages a decoupled architecture with CNText2Sign's pose data for novel direct gesture accuracy assessment. The model employs retrieval augmentation and Cascading Vocabulary Resolution to handle semantic mapping and out-of-vocabulary words and achieves all-scenario production with controllable coordination of gestures and facial expressions via pose-conditioned video synthesis. Concurrently, our Sign Language Translation model SignMST-C employs targeted self-supervised pretraining for dynamic feature capture, achieving new SOTA results on PHOENIX2014-T with BLEU-4 scores up to 32.08. AuraLLM establishes a strong performance baseline on CNText2Sign with a BLEU-4 score of 50.41 under direct evaluation.

cs.CV↗

arXiv2025

Observation of a non-Hermitian supersonic mode on a trapped-ion quantum computer

Quantum computers have long been anticipated to excel in simulating quantum many-body physics. While most previous work has focused on Hermitian physics, we demonstrate the power of variational quantum circuits for resource-efficient simulations of dynamical and equilibrium physics in non-Hermitian systems, revealing new phenomena beyond standard Hermitian quantum machines. Using a variational quantum compilation scheme for fermionic systems, we reduce gate count, save qubits, and eliminate the need for postselection, a major challenge in simulating non-Hermitian dynamics via standard Trotterization. Experimentally, we observed a supersonic mode in the connected density-density correlation function on an $ n = 18 $ fermionic chain after a non-Hermitian, locally interacting quench, which would otherwise be forbidden by the Lieb-Robinson bound in a Hermitian system. Additionally, we investigate sequential quantum circuits generated by tensor networks for ground state preparation, here defined as the eigenstate with the lowest real part eigenvalue, using a variance minimization scheme. Through a trapped-ion implementation on the Quantinuum H1 quantum processor, we accurately capture correlation functions and energies across an exceptional point on a dissipative spin chain up to length $ n = 20 $ using only 3 qubits. Motivated by these advancements, we provide an analytical example demonstrating that simulating single-qubit non-Hermitian dynamics for $Θ(\log(n))$ time from certain initial states is exponentially hard on a quantum computer, offering insights into the opportunities and limitations of using quantum computation for simulating non-Hermitian physics.

quant-ph↗

arXiv2025

ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations

Academic writing requires both coherent text generation and precise citation of relevant literature. Although recent Retrieval-Augmented Generation (RAG) systems have significantly improved factual accuracy in general-purpose text generation, their ability to support professional academic writing remains limited. In this work, we introduce ScholarCopilot, a unified framework designed to enhance existing large language models for generating professional academic articles with accurate and contextually relevant citations. ScholarCopilot dynamically determines when to retrieve scholarly references by generating a retrieval token [RET], which is then used to query a citation database. The retrieved references are fed into the model to augment the generation process. We jointly optimize both the generation and citation tasks within a single framework to improve efficiency. Our model is built upon Qwen-2.5-7B and trained on 500K papers from arXiv. It achieves a top-1 retrieval accuracy of 40.1% on our evaluation dataset, outperforming baselines such as E5-Mistral-7B-Instruct (15.0%) and BM25 (9.8%). On a dataset of 1,000 academic writing samples, ScholarCopilot scores 16.2/25 in generation quality -- measured across relevance, coherence, academic rigor, completeness, and innovation -- significantly surpassing all existing models, including much larger ones like the Retrieval-Augmented Qwen2.5-72B-Instruct. Human studies further demonstrate that ScholarCopilot, despite being a 7B model, significantly outperforms ChatGPT, achieving 100% preference in citation quality and over 70% in overall usefulness.

cs.CL↗

arXiv2025

Stable-Makeup: When Real-World Makeup Transfer Meets Diffusion Model

Current makeup transfer methods are limited to simple makeup styles, making them difficult to apply in real-world scenarios. In this paper, we introduce Stable-Makeup, a novel diffusion-based makeup transfer method capable of robustly transferring a wide range of real-world makeup, onto user-provided faces. Stable-Makeup is based on a pre-trained diffusion model and utilizes a Detail-Preserving (D-P) makeup encoder to encode makeup details. It also employs content and structural control modules to preserve the content and structural information of the source image. With the aid of our newly added makeup cross-attention layers in U-Net, we can accurately transfer the detailed makeup to the corresponding position in the source image. After content-structure decoupling training, Stable-Makeup can maintain content and the facial structure of the source image. Moreover, our method has demonstrated strong robustness and generalizability, making it applicable to varioustasks such as cross-domain makeup transfer, makeup-guided text-to-image generation and so on. Extensive experiments have demonstrated that our approach delivers state-of-the-art (SOTA) results among existing makeup transfer methods and exhibits a highly promising with broad potential applications in various related fields. Code released: https://github.com/Xiaojiu-z/Stable-Makeup

cs.CV↗

arXiv2025

CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer

We present CogVideoX, a large-scale text-to-video generation model based on diffusion transformer, which can generate 10-second continuous videos aligned with text prompt, with a frame rate of 16 fps and resolution of 768 * 1360 pixels. Previous video generation models often had limited movement and short durations, and is difficult to generate videos with coherent narratives based on text. We propose several designs to address these issues. First, we propose a 3D Variational Autoencoder (VAE) to compress videos along both spatial and temporal dimensions, to improve both compression rate and video fidelity. Second, to improve the text-video alignment, we propose an expert transformer with the expert adaptive LayerNorm to facilitate the deep fusion between the two modalities. Third, by employing a progressive training and multi-resolution frame pack technique, CogVideoX is adept at producing coherent, long-duration, different shape videos characterized by significant motions. In addition, we develop an effective text-video data processing pipeline that includes various data preprocessing strategies and a video captioning method, greatly contributing to the generation quality and semantic alignment. Results show that CogVideoX demonstrates state-of-the-art performance across both multiple machine metrics and human evaluations. The model weight of both 3D Causal VAE, Video caption model and CogVideoX are publicly available at https://github.com/THUDM/CogVideo.

cs.CV↗

arXiv2025

Potentials and Limitations of Large-scale, Individual-level Mobile Location Data for Food Acquisition Analysis

Understanding food acquisition is crucial for developing strategies to combat food insecurity, a major public health concern. The emergence of large-scale mobile location data (typically exemplified by GPS data), which captures people's movement over time at high spatiotemporal resolutions, offer a new approach to study this topic. This paper evaluates the potential and limitations of large-scale GPS data for food acquisition analysis through a case study. Using a high-resolution dataset of 286 million GPS records from individuals in Jacksonville, Florida, we conduct a case study to assess the strengths of GPS data in capturing spatiotemporal patterns of food outlet visits while also discussing key limitations, such as potential data biases and algorithmic uncertainties. Our findings confirm that GPS data can generate valuable insights about food acquisition behavior but may significantly underestimate visitation frequency to food outlets. Robustness checks highlight how algorithmic choices-especially regarding food outlet classification and visit identification-can influence research results. Our research underscores the value of GPS data in place-based health studies while emphasizing the need for careful consideration of data coverage, representativeness, algorithmic choices, and the broader implications of study findings.

cs.CY↗

arXiv2025

Solutions with prescribed mass for $L^2$-supercritical NLS equations under Neumann boundary conditions

In this paper, we investigate the following nonlinear Schrödinger equation with Neumann boundary conditions: \begin{equation*} \begin{cases} -Δu+ λu= f(u) & {\rm in} \,~ Ω,\\ \displaystyle\frac{\partial u}{\partial ν}=0 \, &{\rm on}\,~\partial Ω\end{cases} \end{equation*} coupled with a constraint condition: \begin{equation*} \int_Ω|u|^2 dx=c, \end{equation*} where $Ω\subset \mathbb{R}^N(N\ge3)$ denotes a smooth bounded domain, $ν$ represents the unit outer normal vector to $\partial Ω$, $c$ is a positive constant, and $λ$ acts as a Lagrange multiplier. When the nonlinearity $f$ exhibits a general mass supercritical growth at infinity, we establish the existence of normalized solutions, which are not necessarily positive solutions and can be characterized as mountain pass type critical points of the associated constraint functional. Our approach provides a uniform treatment of various nonlinearities, including cases such as $f(u)=|u|^{p-2}u$, $|u|^{q-2}u+ |u|^{p-2}u$, and $-|u|^{q-2}u+|u|^{p-2}u$, where $2<q<2+\frac{4}{N}<p< 2^*$. The result is obtained through a combination of a minimax principle with Morse index information for constrained functionals and a novel blow-up analysis for the NLS equation under Neumann boundary conditions.

math.AP↗

arXiv2025

Standing waves with prescribed mass for NLS equations with Hardy potential in the half-space under Neumman boundary condition

Consider the Neumann problem: \begin{eqnarray*} \begin{cases} &-Δu-\fracμ{|x|^2}u +λu =|u|^{q-2}u+|u|^{p-2}u ~~~\mbox{in}~~\mathbb{R}_+^N,~N\ge3, &\frac{\partial u}{\partial ν}=0 ~~ \mbox{on}~~ \partial\mathbb{R}_+^N \end{cases} \end{eqnarray*} with the prescribed mass: \begin{equation*} \int_{\mathbb{R}_+^N}|u|^2 dx=a>0, \end{equation*} where $\mathbb{R}_+^N$ denotes the upper half-space in $\mathbb{R}^N$, $\frac{1}{|x|^2}$ is the Hardy potential, $2 0$, $ν$ stands for the outward unit normal vector to $\partial \mathbb{R}_+^N$, and $λ$ appears as a Lagrange multiplier. Firstly, by applying Ekeland's variational principle, we establish the existence of normalized solutions that correspond to local minima of the associated energy functional. Furthermore, we find a second normalized solution of mountain pass type by employing a parameterized minimax principle that incorporates Morse index information. Our analysis relies on a Hardy inequality in $H^1(\mathbb{R}_+^N)$, as well as a Pohozaev identity involving the Hardy potential on $\mathbb{R}_+^N$. This work provides a variational framework for investigating the existence of normalized solutions to the Hardy type system within a half-space, and our approach is flexible, allowing it to be adapted to handle more general nonlinearities.

math.AP↗

arXiv2025

EVF-SAM: Early Vision-Language Fusion for Text-Prompted Segment Anything Model

Segment Anything Model (SAM) has attracted widespread attention for its superior interactive segmentation capabilities with visual prompts while lacking further exploration of text prompts. In this paper, we empirically investigate what text prompt encoders (e.g., CLIP or LLM) are good for adapting SAM for referring expression segmentation and introduce the Early Vision-language Fusion-based SAM (EVF-SAM). EVF-SAM is a simple yet effective referring segmentation method which exploits multimodal prompts (i.e., image and text) and comprises a pre-trained vision-language model to generate referring prompts and a SAM model for segmentation. Surprisingly, we observe that: (1) multimodal prompts and (2) vision-language models with early fusion (e.g., BEIT-3) are beneficial for prompting SAM for accurate referring segmentation. Our experiments show that the proposed EVF-SAM based on BEIT-3 can obtain state-of-the-art performance on RefCOCO/+/g for referring expression segmentation and demonstrate the superiority of prompting SAM with early vision-language fusion. In addition, the proposed EVF-SAM with 1.32B parameters achieves remarkably higher performance while reducing nearly 82% of parameters compared to previous SAM methods based on large multimodal models.

cs.CV↗

arXiv2025

EasyControl: Adding Efficient and Flexible Control for Diffusion Transformer

Recent advancements in Unet-based diffusion models, such as ControlNet and IP-Adapter, have introduced effective spatial and subject control mechanisms. However, the DiT (Diffusion Transformer) architecture still struggles with efficient and flexible control. To tackle this issue, we propose EasyControl, a novel framework designed to unify condition-guided diffusion transformers with high efficiency and flexibility. Our framework is built on three key innovations. First, we introduce a lightweight Condition Injection LoRA Module. This module processes conditional signals in isolation, acting as a plug-and-play solution. It avoids modifying the base model weights, ensuring compatibility with customized models and enabling the flexible injection of diverse conditions. Notably, this module also supports harmonious and robust zero-shot multi-condition generalization, even when trained only on single-condition data. Second, we propose a Position-Aware Training Paradigm. This approach standardizes input conditions to fixed resolutions, allowing the generation of images with arbitrary aspect ratios and flexible resolutions. At the same time, it optimizes computational efficiency, making the framework more practical for real-world applications. Third, we develop a Causal Attention Mechanism combined with the KV Cache technique, adapted for conditional generation tasks. This innovation significantly reduces the latency of image synthesis, improving the overall efficiency of the framework. Through extensive experiments, we demonstrate that EasyControl achieves exceptional performance across various application scenarios. These innovations collectively make our framework highly efficient, flexible, and suitable for a wide range of tasks.

cs.CV↗

arXiv2025

Symplectic Optimization for Cross Subcarrier Precoder Design with Channel Smoothing in Massive MIMO-OFDM System

In this paper, we propose a cross subcarrier precoder design (CSPD) for massive multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems. The aim is to maximize the weighted sum-rate (WSR) performance while considering the smoothness of the frequency domain effective channel. To quantify the smoothness of the effective channel, we introduce a delay indicator function to measure the large delay components of the effective channel. An optimization problem is then formulated to balance the WSR performance and the delay indicator function. By appropriately selecting the weight factors in the objective function and the parameters in the delay indicator function, the delay spread of the effective channel can be reduced, thereby enhancing the smoothness of the effective channel. To solve the optimization problem, we apply the symplectic optimization, which achieves faster convergence compared to the gradient descent methods. Simulation results indicate that the proposed algorithm achieves satisfying WSR performance while maintaining the smoothness of the effective channel.

cs.IT↗

arXiv2025

Electric polarization in Chern insulators: Unifying many-body and single-particle approaches

Recently, it has been established that Chern insulators possess an intrinsic two-dimensional electric polarization, despite having gapless edge states and non-localizable Wannier orbitals. This polarization, $\vec{P}_{\text{o}}$, can be defined in a many-body setting from various physical quantities, including dislocation charges, boundary charge distributions, and linear momentum. Importantly, there is a dependence on a choice of real-space origin $\text{o}$ within the unit cell. In contrast, Coh and Vanderbilt extended the single-particle Berry phase definition of polarization to Chern insulators by choosing an arbitrary point in momentum space, $\vec{k}_0$. In this paper, we unify these two approaches and show that when the real-space origin $\text{o}$ and momentum-space point $\vec{k}_0$ are appropriately chosen in relation to each other, the Berry phase and many-body definitions of polarization are equal.

cond-mat.str-el↗

arXiv2025

SECURA: Sigmoid-Enhanced CUR Decomposition with Uninterrupted Retention and Low-Rank Adaptation in Large Language Models

With the rapid development of large language models (LLMs), fully fine-tuning (FT) these models is becoming increasingly infeasible due to high computational demands. Moreover, FT also increases the risk of catastrophic forgetting. As an alternative, Low-Rank Adaptation (LoRA) has been proposed. By fine-tuning only a small subset of parameters, LoRA achieves performance similar to FT while significantly reducing resource requirements. However, since LoRA inherits FT's design, the issue of catastrophic forgetting still remains. To address these limitations, we propose SECURA: Sigmoid-Enhanced CUR Decomposition LoRA, a novel PEFT variant designed to mitigate catastrophic forgetting while improving fine-tuning performance. Our method introduces a novel normalization technique, Sigmoid-based Magnitude Norm (S-MagNorm), which enhances parameter retention and fine-tuning efficiency. SECURA has been evaluated on a diverse range of tasks, including mathematical problem-solving (GSM8K), complex question-answering (CNNDM), translation (NewsDE), and complex multiple-choice reasoning (LogiQA). Experimental results demonstrate that it achieves an average fine-tuning improvement of 3.59% across four MCQ tasks and 2.51% across five QA tasks on Gemma2 2B, Qwen2 1.5B, Qwen2 7B, Llama3 8B, and Llama3.1 8B, outperforming DoRA. Additionally, SECURA demonstrates superior knowledge retention capabilities, achieving state-of-the-art performance in 16 continual learning tests and maintaining more than 70% accuracy on LLMs' basic knowledge compared to Experience Replay (ER), sequential learning (SEQ), EWC, I-LoRA, and CUR-LoRA.

cs.CL↗