arXiv subjects

Yao Zhou

Publications and source records attributed to Yao Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2025

Recognition through Reasoning: Reinforcing Image Geo-localization with Large Vision-Language Models

Previous methods for image geo-localization have typically treated the task as either classification or retrieval, often relying on black-box decisions that lack interpretability. The rise of large vision-language models (LVLMs) has enabled a rethinking of geo-localization as a reasoning-driven task grounded in visual cues. However, two major challenges persist. On the data side, existing reasoning-focused datasets are primarily based on street-view imagery, offering limited scene diversity and constrained viewpoints. On the modeling side, current approaches predominantly rely on supervised fine-tuning, which yields only marginal improvements in reasoning capabilities. To address these challenges, we propose a novel pipeline that constructs a reasoning-oriented geo-localization dataset, MP16-Reason, using diverse social media images. We introduce GLOBE, Group-relative policy optimization for Localizability assessment and Optimized visual-cue reasoning, yielding Bi-objective geo-Enhancement for the VLM in recognition and reasoning. GLOBE incorporates task-specific rewards that jointly enhance localizability assessment, visual-cue reasoning, and geolocation accuracy. Both qualitative and quantitative results demonstrate that GLOBE outperforms state-of-the-art open-source LVLMs on geo-localization tasks, particularly in diverse visual scenes, while also generating more insightful and interpretable reasoning trajectories. The data and code are available at https://github.com/lingli1996/GLOBE.

cs.CV↗

arXiv2025

Entanglement complexification transition driven by a single non-Hermitian impurity

While non-Hermitian bulk systems and their sensitivity to boundary conditions have been extensively studied, how a non-Hermitian boundary affects the entanglement structure of Hermitian critical systems remains largely unexplored. Here we present a fully analytical framework by exactly solving a Hermitian gapless chain with a single non-Hermitian impurity acting as a non-Hermitian boundary. When the entanglement cut is placed at the impurity, we uncover a sharp \emph{entanglement complexification transition}: the logarithmic entanglement entropy retains its scaling form, but the effective central charge evolves from real to complex values, accompanied by a spectral collapse of the correlation matrix. We demonstrate that the real regime follows analytic continuation from a unitary defect conformal field theory (CFT), whereas the complex regime lies entirely beyond this framework. For the latter, we derive an analytical formula in perfect agreement with numerics. Our results reveal that a single non-Hermitian impurity can drive a Hermitian critical system into a nonunitary defect-CFT phase, establishing a rare analytically solvable platform for boundary non-Hermiticity.

quant-ph↗

arXiv2025

Deep Neural Networks for Modeling Astrophysical Nuclear Reacting Flows

In astrophysical simulations, nuclear reacting flows pose computational challenges due to the stiffness of reaction networks. We introduce neural network-based surrogate models using the DeePODE framework to enhance simulation efficiency while maintaining accuracy and robustness. Our method replaces conventional stiff ODE solvers with deep learning models trained through evolutionary Monte Carlo sampling from zero-dimensional simulation data, ensuring generalization across varied thermonuclear and hydrodynamic conditions. Tested on 3-species and 13-species reaction networks, the models achieve $\lesssim 1\%$ accuracy relative to semi-implicit numerical solutions and deliver a $\sim 2.6\times$ speedup on CPUs. A temperature-thresholded deployment strategy ensures stability in extreme conditions, sustaining neural network utilization above 75\% in multi-dimensional simulations. These data-driven surrogates effectively mitigate stiffness constraints, offering a scalable approach for high-fidelity modeling of astrophysical nuclear reacting flows.

astro-ph.IM↗

arXiv2025

Rejection or Inclusion in the Emotion-Identity Dynamics of TikTok Refugees on RedNote

This study examines cross-cultural interactions between Chinese users and self-identified "TikTok Refugees"(foreign users who migrated to RedNote after TikTok's U.S. ban). Based on a dataset of 1,862 posts and 403,054 comments, we use large language model-based sentiment classification and BERT-based topic modelling to explore how both groups engage with the TikTok refugee phenomenon. We analyse what themes foreign users express, how Chinese users respond, how stances (Pro-China, Neutral, Pro-Foreign) shape emotional expression, and how affective responses differ across topics and identities. Results show strong affective asymmetry: Chinese users respond with varying emotional intensities across topics and stances: pride and praise dominate cultural threads, while political discussions elicit high levels of contempt and anger, especially from Pro-China commenters. Pro-Foreign users exhibit the strongest negative emotions across all topics, whereas neutral users express curiosity and joy but still reinforce mainstream discursive norms. Cross-topic comparisons reveal that appearance-related content produces the most emotionally balanced interactions, while politics generates the highest polarization. Our findings reveal distinct emotion-stance structures in Sino-foreign online interactions and offer empirical insights into identity negotiation in transnational digital publics.

cs.SI↗

arXiv2025

Hyperfine Structure of Quantum Entanglement

Quantum entanglement, crucial for understanding quantum many-body systems and quantum gravity, is commonly assessed through various measures such as von Neumann entropy, mutual information, and entanglement contour, each with its inherent advantages and limitations. In this work, we introduce the hyperfine structure of entanglement, which decomposes entanglement contours known as the fine structure into particle-number cumulants. This measure exhibits a set of universal properties with its significance in quantum information science. We apply it across diverse contexts: in Fermi gases, establishing connections to mutual information and interacting conformal field theory; in AdS$_3$/CFT$_2$ correspondence, unveiling finer subregion-subregion duality; and in Chern insulators, distinguishing between different quantum phases, especially topological gapped state and trivial gapped state. Our findings suggest experimental accessibility, offering fresh insights into quantum entanglement across physical systems.

quant-ph↗

arXiv2025

Low-Rank Interconnected Adaptation across Layers

Low-rank adaptation (LoRA) is a widely used parameter-efficient fine-tuning (PEFT) method that learns weight updates $ΔW = AB$ for pretrained weights $W$ through low-rank adapters $A$ and $B$. While LoRA ensures hardware efficiency, its low-rank weight updates limit adaptation performance. In this paper, we propose low-rank interconnected adaptation across layers (Lily), a novel PEFT method that introduces an interconnected framework with locally shared $A$ and globally shared $B$ experts. This structure eliminates redundant per-layer $AB$ pairs, enabling higher-rank $ΔW$ with equal or fewer parameters. To enhance expressiveness, we use data-dependent routers to determine $A$-$B$ interconnections, preventing $B$ experts from converging to the same behavior and improving representational power across domains. Experiments across modalities, architectures, and model sizes demonstrate Lily's superior performance and efficiency. GitHub: https://github.com/yibozhong/lily

cs.CV↗

arXiv2025

Three-dimensional fracton topological orders with boundary Toeplitz braiding

In this paper, we theoretically study a class of 3D non-liquid states that show exotic boundary phenomena in the thermodynamical limit. More concretely, we focus on a class of 3D fracton topological orders formed via stacking 2D twisted $\mathbb{Z}_N$ topologically ordered layers along $z$-direction. Nearby layers are coupled while maintaining translation symmetry along $z$ direction. The effective field theory is given by the infinite-component Chern-Simons (iCS) field theory, with an integer-valued symmetric block-tridiagonal Toeplitz $K$-matrix whose size is thermodynamically large. With open boundary conditions (OBC) along $z$, certain choice of $K$-matrices exhibits exotic boundary ``Toeplitz braiding'', where the mutual braiding phase angle between two anyons at opposite boundaries oscillates and remains non-zero in the thermodynamic limit. In contrast, in trivial case, the mutual braiding phase angle decays exponentially to zero in the thermodynamical limit. As a necessary condition, this phenomenon requires the existence of boundary zero modes in the $K$-matrix spectrum under OBC. We categorize nontrivial $K$-matrices into two distinct types. Each type-I possesses two boundary zero modes, whereas each type-II possesses only one boundary zero mode. Interestingly, the integer-valued Hamiltonian matrix of the familiar 1D ``Su-Schrieffer-Heeger model'' can be used as a non-trivial $K$ matrix. Importantly, since large-gauge-invariance ensures integer quantized $K$-matrix entries, global symmetries are not needed to protect these zero modes. We also present numerical simulation as well as finite size scaling, further confirming the above analytical results. Symmetry fractionalization in iCS field theory is also briefly discussed. Motivated by the present field-theoretical work, it will be interesting to ... ....

cond-mat.str-el↗

arXiv2025

Entanglement scaling behaviors of free fermions on hyperbolic lattices

Recently, tight-binding models on hyperbolic lattices (discretized AdS space) have gained significant attention, leading to hyperbolic band theory and non-Abelian Bloch states. In this paper, we investigate these quantum systems from the perspective of quantum information, focusing particularly on the scaling of entanglement entropy (EE) that has been regarded as a powerful quantum-information probe into exotic phases of matter. It is known that on $d$-dimensional translation-invariant Euclidean lattice, the EE of band insulators scales as an area law ($\sim L^{d-1}$; $L$ is the linear size of the boundary between two subsystems). Meanwhile, the EE of metals (with finite Density-of-State, i.e., DOS) scales as the renowned Gioev-Klich-Widom scaling law ($\sim L^{d-1}\log L$). The appearance of logarithmic divergence, as well as the analytic form of the coefficient $c$ is mathematically controlled by the Widom conjecture of asymptotic behavior of Toeplitz matrices and can be physically understood via the Swingle's argument. However, the hyperbolic lattice, which generalizes translational symmetry, results in inapplicability of these analytic approaches and the potential non-trivial behavior of EE. Here we make an initial attempt through numerical simulation. Remarkably, we find that both cases adhere to the area law, indicating the effect of background hyperbolic geometry that influences quantum entanglement. To achieve the results, we first apply the vertex inflation method to generate hyperbolic lattice on the Poincaré disk, and then apply the Haydock recursion method to compute DOS. Finally, we study the scaling of EE for different bipartitions via exact diagonalization and perform finite-size scaling. We also investigate how the coefficient of the area law is correlated to bulk gap in gapped case and to the DOS in gapless case respectively. Future directions are discussed.

cond-mat.mes-hall↗

arXiv2025

Optimizing Singular Spectrum for Large Language Model Compression

Large language models (LLMs) have demonstrated remarkable capabilities, yet prohibitive parameter complexity often hinders their deployment. Existing singular value decomposition (SVD) based compression methods simply deem singular values as importance scores of decomposed components. However, this importance ordered by singular values does not necessarily correlate with the performance of a downstream task. In this work, we introduce SoCo (Singular spectrum optimization for large language model Compression), a novel compression framework that learns to rescale the decomposed components of SVD in a data-driven manner. Concretely, we employ a learnable diagonal matrix to assign importance scores for singular spectrum and develop a three-stage training process that progressively refines these scores from initial coarse compression to fine-grained sparsification-thereby striking an effective balance between aggressive model compression and performance preservation. Thanks to the learnable singular spectrum, SoCo adaptively prunes components according to the sparsified importance scores, rather than relying on the fixed order of singular values. More importantly, the remaining components with amplified importance scores can compensate for the loss of the pruned ones. Experimental evaluations across multiple LLMs and benchmarks demonstrate that SoCo surpasses the state-of-the-art methods in model compression.

cs.CL↗

arXiv2025

Implementing an information-theoretically secure Byzantine agreement with quantum signed message solution

Byzantine agreement (BA) enables all honest nodes in a decentralized network to reach consensus. In the era of emerging quantum technologies, classical cryptography-based BA protocols face inherent security vulnerabilities. By leveraging the information-theoretic security of keys generated by quantum processing, such as quantum key distribution (QKD), and utilizing the one-time pad (OTP) and one-time universal hashing (OTUH) classical methods proposed in \cite{yin2023QDS}, we propose a quantum signed Byzantine agreement (QSBA) protocol based on the quantum signed message (QSM) scheme. This protocol achieves information-theoretic security using only QKD-shared key resources between network nodes, without requiring quantum entanglement or other advanced quantum resources. Compared to the recently proposed quantum Byzantine agreement (QBA) \cite{weng2023beatingQBA}, our QSBA achieves superior fault tolerance, extending the threshold from nearly 1/2 to an arbitrary number of malicious nodes. Furthermore, our QSBA significantly reduces communication complexity under the same number of malicious nodes. Simulation results in a 5-node twin-field QKD network highlight the efficiency of our protocol, showcasing its potential for secure and resource-efficient consensus in quantum networks.

quant-ph↗

arXiv2025

Collaborative Diffusion Model for Recommender System

Diffusion-based recommender systems (DR) have gained increasing attention for their advanced generative and denoising capabilities. However, existing DR face two central limitations: (i) a trade-off between enhancing generative capacity via noise injection and retaining the loss of personalized information. (ii) the underutilization of rich item-side information. To address these challenges, we present a Collaborative Diffusion model for Recommender System (CDiff4Rec). Specifically, CDiff4Rec generates pseudo-users from item features and leverages collaborative signals from both real and pseudo personalized neighbors identified through behavioral similarity, thereby effectively reconstructing nuanced user preferences. Experimental results on three public datasets show that CDiff4Rec outperforms competitors by effectively mitigating the loss of personalized information through the integration of item content and collaborative signals.

cs.IR↗

arXiv2025

HiMix: Reducing Computational Complexity in Large Vision-Language Models

Benefiting from recent advancements in large language models and modality alignment techniques, existing Large Vision-Language Models(LVLMs) have achieved prominent performance across a wide range of scenarios. However, the excessive computational complexity limits the widespread use of these models in practical applications. We argue that one main bottleneck in computational complexity is caused by the involvement of redundant vision sequences in model computation. This is inspired by a reassessment of the efficiency of vision and language information transmission in the language decoder of LVLMs. Then, we propose a novel hierarchical vision-language interaction mechanism called Hierarchical Vision injection for Mixture Attention (HiMix). In HiMix, only the language sequence undergoes full forward propagation, while the vision sequence interacts with the language at specific stages within each language decoder layer. It is striking that our approach significantly reduces computational complexity with minimal performance loss. Specifically, HiMix achieves a 10x reduction in the computational cost of the language decoder across multiple LVLM models while maintaining comparable performance. This highlights the advantages of our method, and we hope our research brings new perspectives to the field of vision-language understanding. Project Page: https://xuange923.github.io/HiMix

cs.CV↗

arXiv2024

Manga Generation via Layout-controllable Diffusion

Generating comics through text is widely studied. However, there are few studies on generating multi-panel Manga (Japanese comics) solely based on plain text. Japanese manga contains multiple panels on a single page, with characteristics such as coherence in storytelling, reasonable and diverse page layouts, consistency in characters, and semantic correspondence between panel drawings and panel scripts. Therefore, generating manga poses a significant challenge. This paper presents the manga generation task and constructs the Manga109Story dataset for studying manga generation solely from plain text. Additionally, we propose MangaDiffusion to facilitate the intra-panel and inter-panel information interaction during the manga generation process. The results show that our method particularly ensures the number of panels, reasonable and diverse page layouts. Based on our approach, there is potential to converting a large amount of textual stories into more engaging manga readings, leading to significant application prospects.

cs.CV↗

arXiv2024

Quantum entanglement and non-Hermiticity in free-fermion systems

This topical review article reports rapid progress on the generalization and application of entanglement in non-Hermitian free-fermion quantum systems. We begin by examining the realization of non-Hermitian quantum systems through the Lindblad master equation, alongside a review of typical non-Hermitian free-fermion systems that exhibit unique features. A pedagogical discussion is provided on the relationship between entanglement quantities and the correlation matrix in Hermitian systems. Building on this foundation, we focus on how entanglement concepts are extended to non-Hermitian systems from their Hermitian free-fermion counterparts, with a review of the general properties that emerge. Finally, we highlight various concrete studies, demonstrating that entanglement entropy remains a powerful diagnostic tool for characterizing non-Hermitian physics. The entanglement spectrum also reflects the topological characteristics of non-Hermitian topological systems, while unique non-Hermitian entanglement behaviors are also discussed. The review is concluded with several future directions. Through this review, we hope to provide a useful guide for researchers who are interested in entanglement in non-Hermitian quantum systems.

quant-ph↗

arXiv2024

Rethinking Low-Rank Adaptation in Vision: Exploring Head-Level Responsiveness across Diverse Tasks

Low-rank adaptation (LoRA) has shifted the paradigm of adapting pre-trained Vision Transformers (ViT), achieving great efficiency by updating only a subset of tailored parameters to approximate weight updates. However, the multi-head design of the self-attention mechanism, with the heads working in parallel in the computation flow, exhibiting similar visual patterns and requiring update over all of them, incurs unnecessary storage and computational overhead. In this paper, we propose Head-level responsiveness tuning for low-rank adaptation (Heart-LoRA). The proposed method explores redundancy among the heads and selectively activates task-responsive heads, thus enabling fine-grained head-level tuning. Additionally, given the different responsiveness of heads to diverse visual tasks, our proposed method dynamically activates a subset of the approximated heads that are tailored to the current task. Experimental results show that Heart-LoRA yields superior performance over state-of-the-art PETL approaches on visual adaptation benchmark datasets.

cs.CV↗

arXiv2024

Entanglement Fractalization

We numerically explore the interplay of fractal geometry and quantum entanglement by analyzing the von Neumann entropy (known as entanglement entropy) and the entanglement contour in the scaling limit. Adopting quadratic fermionic models on Sierpinski carpet, we uncover intriguing findings. For gapless ground states exhibiting a finite density of states at the chemical potential, we reveal a super-area law characterized by the presence of a logarithmic correction for area law in the scaling of entanglement entropy. This extends the well-established super-area law observed on translationally invariant Euclidean lattices where the Gioev-Klich-Widom conjecture regarding the asymptotic behavior of Toeplitz matrices holds significant influence. Furthermore, different from the fractal structure of the lattice, we observe the emergence of a novel self-similar and universal pattern termed an ``entanglement fractal'' in the entanglement contour data as we approach the scaling limit. Remarkably, this pattern bears resemblance to intricate Chinese paper-cutting designs. We provide general rules to artificially generate this fractal, offering insights into the universal scaling of entanglement entropy. Meanwhile, as the direct consequence of the entanglement fractal and beyond a single scaling behavior of entanglement contour in translation-invariant systems, we identify two distinct scaling behaviors in the entanglement contour of fractal systems. For gapped ground states, we observe that the entanglement entropy adheres to a generalized area law, with its dependence on the Hausdorff dimension of the boundary between complementary subsystems.

quant-ph↗

arXiv2024

Pear: Pruning and Sharing Adapters in Visual Parameter-Efficient Fine-Tuning

Adapters have been widely explored to alleviate computational and storage costs when fine-tuning pretrained foundation models. However, the adapter itself can exhibit redundancy, leading to unnecessary storage overhead and inferior performance. In this paper, we propose Prune and Share (Pear), a novel adapter-pruning framework for efficient fine-tuning of pretrained visual foundation models. Specifically, we prune certain adapters and share the more important unpruned ones with positions where adapters are pruned, allowing continual adaptation at these positions after pruning. Additionally, a knowledge checkpoint strategy is introduced, which preserves the information of the pruned adapters and further boosts performance. Experimental results on visual adaptation benchmark validate the effectiveness and efficiency of the proposed Pear comparing to other competitive methods. Code is in https://github.com/yibozhong/pear.

cs.CV↗

arXiv2024

Benign saturation of ideal ballooning instability in a high-performance stellarator

To examine the robustness of the designed 5% $β$-limit for high-performance operation in the W7-X stellarator, we undertake nonlinear magnetohydrodynamic (MHD) simulations of pressure-driven instabilities using the M3D-$C^1$ code. Consistent with linear analyses, ideal ballooning instabilities occur as $β$ exceeds 5% in the standard configuration. Nonetheless, the modes saturate nonlinearly at relatively low levels without triggering large-scale crashes, even though confinement degradation worsens as $β$ increases. In contrast, in an alternative configuration with nearly zero magnetic shear, ideal interchange modes induce a pressure crash at $β= 1\%$. These results suggest that the standard W7-X configuration might have a soft $β$-limit that is fairly immune to major MHD events, which enhances the appeal of the stellarator approach to steady-state fusion reactors.

physics.plasm-ph↗