arXiv subjects

Yang Zhao

Publications and source records attributed to Yang Zhao.

At least 91 records · Page 5Linked to original sources

arXiv2026

SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training

Recent advances in diffusion-based video restoration (VR) demonstrate significant improvement in visual quality, yet yield a prohibitive computational cost during inference. While several distillation-based approaches have exhibited the potential of one-step image restoration, extending existing approaches to VR remains challenging and underexplored, particularly when dealing with high-resolution video in real-world settings. In this work, we propose a one-step diffusion-based VR model, termed as SeedVR2, which performs adversarial VR training against real data. To handle the challenging high-resolution VR within a single step, we introduce several enhancements to both model architecture and training procedures. Specifically, an adaptive window attention mechanism is proposed, where the window size is dynamically adjusted to fit the output resolutions, avoiding window inconsistency observed under high-resolution VR using window attention with a predefined window size. To stabilize and improve the adversarial post-training towards VR, we further verify the effectiveness of a series of losses, including a proposed feature matching loss without significantly sacrificing training efficiency. Extensive experiments show that SeedVR2 can achieve comparable or even better performance compared with existing VR approaches in a single step.

cs.CV↗

arXiv2026

VaseVQA: Multimodal Agent and Benchmark for Ancient Greek Pottery

Understanding cultural heritage artifacts such as ancient Greek pottery requires expert-level reasoning that remains challenging for current MLLMs due to limited domain-specific data. We introduce VaseVQA, a benchmark of 31,773 images and 67,614 question-answer pairs across seven expert-defined categories, enabling systematic evaluation of expert-level cultural heritage understanding. Using this dataset, we explore effective training strategies for domain-specific reasoning. While supervised fine-tuning improves adaptation to domain knowledge, it struggles with deeper reasoning tasks. We propose VaseVL, which augments SFT with reinforcement learning using verifiable rewards. Experiments show that VaseVL consistently outperforms supervised baselines, especially on reasoning-intensive questions, highlighting the value of targeted reinforcement learning for cultural heritage visual question answering. Our code and dataset will be released at https://github.com/AIGeeksGroup/VaseVQA.

cs.CV↗

arXiv2026

SKANet: A Cognitive Dual-Stream Framework with Adaptive Modality Fusion for Robust Compound GNSS Interference Classification

As the electromagnetic environment becomes increasingly complex, Global Navigation Satellite Systems (GNSS) face growing threats from sophisticated jamming interference. Although Deep Learning (DL) effectively identifies basic interference, classifying compound interference remains difficult due to the superposition of diverse jamming sources. Existing single-domain approaches often suffer from performance degradation because transient burst signals and continuous global signals require conflicting feature extraction scales. We propose the Selective Kernel and Asymmetric convolution Network(SKANet), a cognitive deep learning framework built upon a dual-stream architecture that integrates Time-Frequency Images (TFIs) and Power Spectral Density (PSD). Distinct from conventional fusion methods that rely on static receptive fields, the proposed architecture incorporates a Multi-Branch Selective Kernel (SK) module combined with Asymmetric Convolution Blocks (ACBs). This mechanism enables the network to dynamically adjust its receptive fields, acting as an adaptive filter that simultaneously captures micro-scale transient features and macro-scale spectral trends within entangled compound signals. To complement this spatial-temporal adaptation, a Squeeze-and-Excitation (SE) mechanism is integrated at the fusion stage to adaptively recalibrate the contribution of heterogeneous features from each modality. Evaluations on a dataset of 405,000 samples demonstrate that SKANet achieves an overall accuracy of 96.99\%, exhibiting superior robustness for compound jamming classification, particularly under low Jamming-to-Noise Ratio (JNR) regimes.

cs.CV↗

arXiv2026

PhyG-MoE: A Physics-Guided Mixture-of-Experts Framework for Energy-Efficient GNSS Interference Recognition

Complex electromagnetic interference increasingly compromises Global Navigation Satellite Systems (GNSS), threatening the reliability of Space-Air-Ground Integrated Networks (SAGIN). Although deep learning has advanced interference recognition, current static models suffer from a \textbf{fundamental limitation}: they impose a fixed computational topology regardless of the input's physical entropy. This rigidity leads to severe resource mismatch, where simple primitives consume the same processing cost as chaotic, saturated mixtures. To resolve this, this paper introduces PhyG-MoE (Physics-Guided Mixture-of-Experts), a framework designed to \textbf{dynamically align model capacity with signal complexity}. Unlike static architectures, the proposed system employs a spectrum-based gating mechanism that routes signals based on their spectral feature entanglement. A high-capacity TransNeXt expert is activated on-demand to disentangle complex features in saturated scenarios, while lightweight experts handle fundamental signals to minimize latency. Evaluations on 21 jamming categories demonstrate that PhyG-MoE achieves an overall accuracy of 97.58\%. By resolving the intrinsic conflict between static computing and dynamic electromagnetic environments, the proposed framework significantly reduces computational overhead without performance degradation, offering a viable solution for resource-constrained cognitive receivers.

cs.CV↗

arXiv2025

HiSciBench: A Hierarchical Multi-disciplinary Benchmark for Scientific Intelligence from Reading to Discovery

The rapid advancement of large language models (LLMs) and multimodal foundation models has sparked growing interest in their potential for scientific research. However, scientific intelligence encompasses a broad spectrum of abilities ranging from understanding fundamental knowledge to conducting creative discovery, and existing benchmarks remain fragmented. Most focus on narrow tasks and fail to reflect the hierarchical and multi-disciplinary nature of real scientific inquiry. We introduce \textbf{HiSciBench}, a hierarchical benchmark designed to evaluate foundation models across five levels that mirror the complete scientific workflow: \textit{Scientific Literacy} (L1), \textit{Literature Parsing} (L2), \textit{Literature-based Question Answering} (L3), \textit{Literature Review Generation} (L4), and \textit{Scientific Discovery} (L5). HiSciBench contains 8,735 carefully curated instances spanning six major scientific disciplines, including mathematics, physics, chemistry, biology, geography, and astronomy, and supports multimodal inputs including text, equations, figures, and tables, as well as cross-lingual evaluation. Unlike prior benchmarks that assess isolated abilities, HiSciBench provides an integrated, dependency-aware framework that enables detailed diagnosis of model capabilities across different stages of scientific reasoning. Comprehensive evaluations of leading models, including GPT-5, DeepSeek-R1, and several multimodal systems, reveal substantial performance gaps: while models achieve up to 69\% accuracy on basic literacy tasks, performance declines sharply to 25\% on discovery-level challenges. HiSciBench establishes a new standard for evaluating scientific Intelligence and offers actionable insights for developing models that are not only more capable but also more reliable. The benchmark will be publicly released to facilitate future research.

cs.AI↗

arXiv2025

Entanglement wedge cross section triangle information and holographic entanglement of assistance

We identify a non-negative and upper-bounded entanglement signal in holography which is defined as a combination of entanglement wedge cross sections (EWCS) for a tripartite mixed state $ABE$: $\mathrm{EI}_Δ(A:B|E) = \mathrm{EWCS}(A:EB) + \mathrm{EWCS}(B:EA) - \mathrm{EWCS}(E:AB)$. This quantity is an analogue of conditional mutual information (CMI) and shares similar mathematical structures in both quantum information theory and holography. We show that CMI is upper bounded by a quantum information quantity, the entanglement of assistance, which quantifies the entanglement that can be generated between two parties $A$ and $B$, given assistance from a third party $E$. We prove that $\mathrm{EI}_Δ$ is also upper bounded by the entanglement of assistance in the canonical purification state. We analyze its upper bound by maximizing $\mathrm{EI}_Δ(A:B|E)$ over all configurations of the auxiliary subsystem $E$ in AdS$_3$/CFT$_2$. The maximized $\mathrm{EI}_Δ$ displays a rich phase structure governed by the cross ratio $X_{AB}$: it vanishes below a critical threshold and, beyond a second phase transition point, saturates the bound of entanglement of assistance. We comment on the interpretation of $\mathrm{EI}_Δ$ as characterizing the assisted bipartite quantum entanglement between $A$ and $B$ with the help of $E$.

hep-th↗

arXiv2025

Holographic multipartite entanglement structures in IR modified geometries

We investigate how IR modifications of the bulk geometry reshape long-range multipartite entanglement on the boundary in holography. We modify the IR geometries in two opposite directions: spherical modifications that enhance long-range entanglement and hyperbolic modifications that suppress them. We utilize various multipartite entanglement measures/signals to analyze the multipartite entanglement structures. These measures/signals are combinations of entanglement entropy, multi-entropy, entanglement wedge cross sections (EWCS) and multi-EWCS. Our results reveal that in the extremal limits of these two geometric modifications, the multipartite entanglement structures exhibit starkly contrasting behaviors: various measures saturate either their theoretical upper or lower bounds in the respective geometries. This demonstrates that IR deformations provide a practical holographic framework for realizing extremal entanglement regimes. Moreover, it serves as an effective tool for studying quantum marginal problems in holography. Finally, by observing how different measures respond to these engineered geometries, we gain clarifying insights into the specific types of multipartite entanglement that each measure/signal is particularly sensitive to.

hep-th↗

arXiv2025

New Paradigm for Integrated Sensing and Communication with Rydberg Atomic Receiver

The RYDberg Atomic Receiver (RYDAR) has been demonstrated to surmount the limitation on both the sensitivity and operating bandwidth of the classical electronic counterpart, which can theoretically detect indiscernible electric signals below -174 dBm/Hz with optical measurement through Rydberg-state atoms. Such miracle has established a new quantum-based paradigm for communications and sensing, which motivates a revolution of the transceiver design philosophies to fully unleash the potential of RYDAR towards next-generation networks. Against this background, this article provides a thorough investigation of Rydberg atomic communications and sensing from theory to hardware implementations. Specifically, we highlight the great opportunities from the hybridization between the RYDAR and the cutting-edge integrated sensing and communication (ISAC), followed by essential preliminaries of the quantum-based receiver. Then we propose a theoretical framework for broadband ISAC based on RYDAR, demonstrated by the proof-of-concept experiments. Afterwards, the enabling technologies for the ISAC framework are explored ranging from channel characterization, waveform design to array-based receiver configurations, where the open problems are also summarized. Finally, the future applications of RYDAR-based ISAC are envisioned, indicating its significant potential for both civilian and military purposes.

quant-ph↗

arXiv2025

Physical Computing at the Data Processing Inequality Limit

Wave-physics-based intelligent sensing has driven multidisciplinary applications from smart industries to decision-making systems. Traditional sensing paradigms transform physical waveforms into human-understandable intermediate representations through preprocessing. Such transformations inherently cause information loss owing to data processing inequality (DPI). Here, we established a theoretical framework for physical computing at the DPI upper limit. Physical computing avoids information loss during preprocessing by directly extracting information from physical waveforms, achieving the theoretical maximum of accessible information as determined by the DPI. Furthermore, physical computing comprehensively utilizes multiple dimensions of physical waveforms, thereby enhancing the upper limit of information capture capability. Electromagnetic sensing experiments have demonstrated that physical computing can achieve 100% sensing accuracy, substantially outperforming traditional sensing paradigms. The proposed theoretical framework of physical computing offers a promising path towards enhancing the information-capture capability of next-generation intelligent sensing systems.

physics.app-ph↗

arXiv2025

Leveraging Reference Documents for Zero-Shot Ranking via Large Language Models

Large Language Models (LLMs) have demonstrated exceptional performance in the task of text ranking for information retrieval. While Pointwise ranking approaches offer computational efficiency by scoring documents independently, they often yield biased relevance estimates due to the lack of inter-document comparisons. In contrast, Pairwise methods improve ranking accuracy by explicitly comparing document pairs, but suffer from substantial computational overhead with quadratic complexity ($O(n^2)$). To address this tradeoff, we propose \textbf{RefRank}, a simple and effective comparative ranking method based on a fixed reference document. Instead of comparing all document pairs, RefRank prompts the LLM to evaluate each candidate relative to a shared reference anchor. By selecting the reference anchor that encapsulates the core query intent, RefRank implicitly captures relevance cues, enabling indirect comparison between documents via this common anchor. This reduces computational cost to linear time ($O(n)$) while importantly, preserving the advantages of comparative evaluation. To further enhance robustness, we aggregate multiple RefRank outputs using a weighted averaging scheme across different reference choices. Experiments on several benchmark datasets and with various LLMs show that RefRank significantly outperforms Pointwise baselines and could achieve performance at least on par with Pairwise approaches with a significantly lower computational cost.

cs.IR↗

arXiv2025

A Fine Evaluation Method for Cube Copying Test for Early Detection of Alzheimer's Disease

Background: Impairment of visual spatial cognitive function is the most common early clinical manifestation of Alzheimer's Disease (AD). When the Montreal Cognitive Assessment (MoCA) uses the "0/1" binary method ("pass/fail") to evaluate the visual spatial cognitive ability represented by the Cube Copying Test(CCT), the elder with less formal education generally score 0 point, resulting in serious bias in the evaluation results. Therefore, this study proposes a fine evaluation method for CCT based on dynamic handwriting feature extraction of DH-SCSM-BLA. method : The Cogni-CareV3.0 software independently developed by our team was used to collect dynamic handwriting data of CCT. Then, the spatial and motion features of segmented dynamic handwriting were extracted, and feature matrix with unequal dimensions were normalized. Finally, a bidirectional long short-term memory network model combined with attention mechanism (BiLSTM-Attention) was adopted for classification. Result: The experimental results showed that: The proposed method has significant superiority compared to similar studies, with a classification accuracy of 86.69%. The distribution of cube drawing ability scores has significant regularity for three aspects such as MCI patients and healthy control group, age, and levels of education. It was also found that score for each cognitive task including cube drawing ability score is negatively correlated with age. Score for each cognitive task including cube drawing ability score, but positively correlated with levels of education significantly. Conclusion: This study provides a relatively objective and comprehensive evaluation method for early screening and personalized intervention of visual spatial cognitive impairment.

cs.LG↗

arXiv2025

PINNsFailureRegion Localization and Refinement through White-box AdversarialAttack

Physics-informed neural networks (PINNs) have shown great promise in solving partial differential equations (PDEs). However, vanilla PINNs often face challenges when solving complex PDEs, especially those involving multi-scale behaviors or solutions with sharp or oscillatory characteristics. To precisely and adaptively locate the critical regions that fail in the solving process we propose a sampling strategy grounded in white-box adversarial attacks, referred to as WbAR. WbAR search for failure regions in the direction of the loss gradient, thus directly locating the most critical positions. WbAR generates adversarial samples in a random walk manner and iteratively refines PINNs to guide the model's focus towards dynamically updated critical regions during training. We implement WbAR to the elliptic equation with multi-scale coefficients, Poisson equation with multi-peak solutions, high-dimensional Poisson equations, and Burgers equation with sharp solutions. The results demonstrate that WbAR can effectively locate and reduce failure regions. Moreover, WbAR is suitable for solving complex PDEs, since locating failure regions through adversarial attacks is independent of the size of failure regions or the complexity of the distribution.

cs.LG↗

arXiv2025

Re-examining the boundary conditions in modelling SAW-driven acoustofluidic streaming

Numerical simulations of surface acoustic wave (SAW)-induced acoustic streaming are highly sensitive to the choice of second-order boundary conditions. This study systematically compares the no-slip (NS) and Stokes slip (SD) boundary conditions through different numerical approaches. Two- and three-dimensional simulations based on the Reynolds stress method are performed for standing SAW and travelling SAW devices. Results are validated against particle image velocimetry measurements of streaming patterns and velocities. We show that the SD condition yields Lagrangian velocity fields in significantly better agreement with experiments than the NS condition, accurately capturing vortex number, rotation direction, and amplitude across varying device geometries and operating conditions. In contrast, the NS condition overpredicts velocities by 1-2 orders of magnitude and often fails to reproduce experimentally observed vortex structures. These findings highlight the essential role of the Stokes drift boundary condition in modelling acoustic streaming and provide clear guidance for its use in future simulations of SAW-based acoustofluidic systems.

physics.flu-dyn↗

arXiv2025

Tunable Multistage Refrigeration via Geometrically Frustrated Triangular Lattice Antiferromagnet for Space Cooling

Low-temperature refrigeration technology constitutes a crucial component in space exploration. The small-scale, low-vibration Stirling-type pulse tube refrigerators hold significant application potential for space cooling. However, the efficient operation of current Stirling-type pulse tube cryocoolers in space cooling applications remains challenging due to the rapid decay of the heat capacity of regenerative materials below 10 K. This study adopts a novel material strategy: using a novel high-spin S = 7/2 magnetic regenerative material, Gd2O2Se, we construct a multistage tunable regenerative material structure to achieve an efficient cooling approach to the liquid helium temperature range. Under substantial geometric frustration from a double-layered triangular lattice, it exhibits two-step specific heat transition peaks at 6.22 K and 2.11 K, respectively. Its ultrahigh specific heat and broad two-step transition temperature range effectively bridge the gap between commercially used high-heat-capacity materials. Experimental verification shows that when Gd2O2Se is combined with Er3Ni and HoCu2 in the Stirling-type pulse tube cryocooler, the cooling efficiency of the pulse tube increases by 66.5 % at 7 K, and the minimum achievable temperature reaches 5.85 K. These results indicate that Gd2O2Se is an ideal magnetic regenerative material for space cooling

cond-mat.mtrl-sci↗

arXiv2025

Personalized Decision Modeling: Utility Optimization or Textualized-Symbolic Reasoning

Decision-making models for individuals, particularly in high-stakes scenarios like vaccine uptake, often diverge from population optimal predictions. This gap arises from the uniqueness of the individual decision-making process, shaped by numerical attributes (e.g., cost, time) and linguistic influences (e.g., personal preferences and constraints). Developing upon Utility Theory and leveraging the textual-reasoning capabilities of Large Language Models (LLMs), this paper proposes an Adaptive Textual-symbolic Human-centric Reasoning framework (ATHENA) to address the optimal information integration. ATHENA uniquely integrates two stages: First, it discovers robust, group-level symbolic utility functions via LLM-augmented symbolic discovery; Second, it implements individual-level semantic adaptation, creating personalized semantic templates guided by the optimal utility to model personalized choices. Validated on real-world travel mode and vaccine choice tasks, ATHENA consistently outperforms utility-based, machine learning, and other LLM-based models, lifting F1 score by at least 6.5% over the strongest cutting-edge models. Further, ablation studies confirm that both stages of ATHENA are critical and complementary, as removing either clearly degrades overall predictive performance. By organically integrating symbolic utility modeling and semantic adaptation, ATHENA provides a new scheme for modeling human-centric decisions. The project page can be found at https://yibozh.github.io/Athena.

cs.AI↗

arXiv2025

HiRA: A Hierarchical Reasoning Framework for Decoupled Planning and Execution in Deep Search

Complex information needs in real-world search scenarios demand deep reasoning and knowledge synthesis across diverse sources, which traditional retrieval-augmented generation (RAG) pipelines struggle to address effectively. Current reasoning-based approaches suffer from a fundamental limitation: they use a single model to handle both high-level planning and detailed execution, leading to inefficient reasoning and limited scalability. In this paper, we introduce HiRA, a hierarchical framework that separates strategic planning from specialized execution. Our approach decomposes complex search tasks into focused subtasks, assigns each subtask to domain-specific agents equipped with external tools and reasoning capabilities, and coordinates the results through a structured integration mechanism. This separation prevents execution details from disrupting high-level reasoning while enabling the system to leverage specialized expertise for different types of information processing. Experiments on four complex, cross-modal deep search benchmarks demonstrate that HiRA significantly outperforms state-of-the-art RAG and agent-based systems. Our results show improvements in both answer quality and system efficiency, highlighting the effectiveness of decoupled planning and execution for multi-step information seeking tasks. Our code is available at https://github.com/ignorejjj/HiRA.

cs.AI↗

arXiv2025

Seeing Symbols, Missing Cultures: Probing Vision-Language Models' Reasoning on Fire Imagery and Cultural Meaning

Vision-Language Models (VLMs) often appear culturally competent but rely on superficial pattern matching rather than genuine cultural understanding. We introduce a diagnostic framework to probe VLM reasoning on fire-themed cultural imagery through both classification and explanation analysis. Testing multiple models on Western festivals, non-Western traditions, and emergency scenes reveals systematic biases: models correctly identify prominent Western festivals but struggle with underrepresented cultural events, frequently offering vague labels or dangerously misclassifying emergencies as celebrations. These failures expose the risks of symbolic shortcuts and highlight the need for cultural evaluation beyond accuracy metrics to ensure interpretable and fair multimodal systems.

cs.CV↗

arXiv2025

Movable Antenna Enhanced Federated Fine-Tuning of Large Language Models via Hybrid Client Selection Optimization

Federated fine-tuning of large language models (LLMs) over bandwidth-limited 6G links must meet strict round-time and energy budgets. Analog over-the-air (OTA) aggregation reduces uplink cost but is sensitive to fading and interference, which distort the aggregated gradient. We consider a two-phase workflow (centralized pre-training followed by federated fine-tuning) where the base station uses a movable-antenna (MA) array. In each round, MA element positions and the receive/transmit beamformers are adjusted under minimum-spacing constraints to reshape the channel and improve OTA aggregation without increasing user transmit power. We formulate a mixed-integer, nonconvex resource-allocation problem that jointly selects clients and optimizes the number of global rounds, CPU frequencies, mini-batch sizes, MA positions, and analog weights under end-to-end latency and energy limits. A successive convex approximation-penalty dual decomposition (SCA-PDD) routine alternates convex updates with oblique-manifold beamforming and spacing-aware MA placement. Experiments on OpenLLaMA-v2 (3B) with LoRA and 4-bit quantization on Alpaca and Dolly (10 clients) attain round-30 validation perplexities as low as 2.94 (Alpaca, K=1) and 4.62 (Dolly, K=1). Relative to the strongest non-MA baseline at the same concurrency, this corresponds to 17.4 percent (Alpaca, K=1) and 54.4 percent (Dolly, K=1) lower perplexity; at K=2 the reductions are 14.2 percent (Alpaca) and 13.7 percent (Dolly). Participation fairness also improves across all uplink concurrencies K in {1,2,4,8}, with the largest margins when fewer clients transmit per round.

eess.SP↗