arXiv subjects

Kun Ding

Publications and source records attributed to Kun Ding.

At least 37 records · Page 2Linked to original sources

arXiv2026

DSFC-Net: A Dual-Encoder Spatial and Frequency Co-Awareness Network for Rural Road Extraction

Accurate extraction of rural roads from high-resolution remote sensing imagery is essential for infrastructure planning and sustainable development. However, this task presents unique challenges in rural settings due to several factors. These include high intra-class variability and low inter-class separability from diverse surface materials, frequent vegetation occlusions that disrupt spatial continuity, and narrow road widths that exacerbate detection difficulties. Existing methods, primarily optimized for structured urban environments, often underperform in these scenarios as they overlook such distinctive characteristics. To address these challenges, we propose DSFC-Net, a dual-encoder framework that synergistically fuses spatial and frequency-domain information. Specifically, a CNN branch is employed to capture fine-grained local road boundaries and short-range continuity, while a novel Spatial-Frequency Hybrid Transformer (SFT) is introduced to robustly model global topological dependencies against vegetation occlusions. Distinct from standard attention mechanisms that suffer from frequency bias, the SFT incorporates a Cross-Frequency Interaction Attention (CFIA) module that explicitly decouples high- and low-frequency information via a Laplacian Pyramid strategy. This design enables the dynamic interaction between spatial details and frequency-aware global contexts, effectively preserving the connectivity of narrow roads. Furthermore, a Channel Feature Fusion Module (CFFM) is proposed to bridge the two branches by adaptively recalibrating channel-wise feature responses, seamlessly integrating local textures with global semantics for accurate segmentation. Comprehensive experiments on the WHU-RuR+, DeepGlobe, and Massachusetts datasets validate the superiority of DSFC-Net over state-of-the-art approaches.

cs.CV↗

arXiv2026

Geometry-induced Exceptional Point Detached from Fermi Arcs

Exceptional points (EPs), ubiquitous non-Hermitian degeneracies, are central features in band structures where non-Hermitian Fermi arcs connect EPs and eigenvalue knots encircle them. Under open boundary conditions (OBCs), non-Hermitian skin effects enforce complex momenta and yield non-Bloch band structures, introducing EPs unique to OBCs whose origins depend on boundary-driven mechanisms. Here, we reveal both theoretically and experimentally that geometry itself can induce such non-Bloch EPs in a reciprocal non-Hermitian Lieb lattice supporting geometry-dependent skin effects. By analyzing non-Bloch band structures, we find that geometry-induced EPs correspond to saddle points rather than branch points. Branch points, even while carrying OBC eigenenergies, do not yield EPs but manifest as Whitney cusps, a characteristic type of geometric singularity, and Fermi arcs connecting them remain crucial in determining eigenvalue knots. Our measurements of these knots confirm that geometry-induced EPs are detached from the branch points of Fermi arcs, contrasting with their unified counterparts in Bloch systems. Our results establish geometry as an additional degree of freedom for engineering EP-based devices and reveal its fundamental role in shaping non-Bloch band structures.

cond-mat.other↗

arXiv2025

PDE-Agent: A toolchain-augmented multi-agent framework for PDE solving

Solving Partial Differential Equations (PDEs) is a cornerstone of engineering and scientific research. Traditional methods for PDE solving are cumbersome, relying on manual setup and domain expertise. While Physics-Informed Neural Network (PINNs) introduced end-to-end neural network-based solutions, and frameworks like DeepXDE further enhanced automation, these approaches still depend on expert knowledge and lack full autonomy. In this work, we frame PDE solving as tool invocation via LLM-driven agents and introduce PDE-Agent, the first toolchain-augmented multi-agent collaboration framework, inheriting the reasoning capacity of LLMs and the controllability of external tools and enabling automated PDE solving from natural language descriptions. PDE-Agent leverages the strengths of multi-agent and multi-tool collaboration through two key innovations: (1) A Prog-Act framework with graph memory for multi-agent collaboration, which enables effective dynamic planning and error correction via dual-loop mechanisms (localized fixes and global revisions). (2) A Resource-Pool integrated with a tool-parameter separation mechanism for multi-tool collaboration. This centralizes the management of runtime artifacts and resolves inter-tool dependency gaps in existing frameworks. To validate and evaluate this new paradigm for PDE solving , we develop PDE-Bench, a multi-type PDE Benchmark for agent-based tool collaborative solving, and propose multi-level metrics for assessing tool coordination. Evaluations verify that PDE-Agent exhibits superior applicability and performance in complex multi-step, cross-step dependent tasks. This new paradigm of toolchain-augmented multi-agent PDE solving will further advance future developments in automated scientific computing. Our source code and dataset will be made publicly available.

cs.AI↗

arXiv2025

Efficient Redundancy Reduction for Open-Vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation (OVSS) is an open-world task that aims to assign each pixel within an image to a specific class defined by arbitrary text descriptions. While large-scale vision-language models have shown remarkable open-vocabulary capabilities, their image-level pretraining limits effectiveness on pixel-wise dense prediction tasks like OVSS. Recent cost-based methods narrow this granularity gap by constructing pixel-text cost maps and refining them via cost aggregation mechanisms. Despite achieving promising performance, these approaches suffer from high computational costs and long inference latency. In this paper, we identify two major sources of redundancy in the cost-based OVSS framework: redundant information introduced during cost maps construction and inefficient sequence modeling in cost aggregation. To address these issues, we propose ERR-Seg, an efficient architecture that incorporates Redundancy-Reduced Hierarchical Cost maps (RRHC) and Redundancy-Reduced Cost Aggregation (RRCA). Specifically, RRHC reduces redundant class channels by customizing a compact class vocabulary for each image and integrates hierarchical cost maps to enrich semantic representation. RRCA alleviates computational burden by performing both spatial-level and class-level sequence reduction before aggregation. Overall, ERR-Seg results in a lightweight structure for OVSS, characterized by substantial memory and computational savings without compromising accuracy. Compared to previous state-of-the-art methods on the ADE20K-847 benchmark, ERR-Seg improves performance by $5.6\%$ while achieving a 3.1$\times$ speedup.

cs.CV↗

arXiv2025

Fault Diagnosis and Quantification for Photovoltaic Arrays based on Differentiable Physical Models

Accurate fault diagnosis and quantification are essential for the reliable operation and intelligent maintenance of photovoltaic (PV) arrays. However, existing fault quantification methods often suffer from limited efficiency and interpretability. To address these challenges, this paper proposes a novel fault quantification approach for PV strings based on a differentiable fast fault simulation model (DFFSM). The proposed DFFSM accurately models I-V characteristics under multiple faults and provides analytical gradients with respect to fault parameters. Leveraging this property, a gradient-based fault parameters identification (GFPI) method using the Adahessian optimizer is developed to efficiently quantify partial shading, short-circuit, and series-resistance degradation. Experimental results on both simulated and measured I-V curves demonstrate that the proposed GFPI achieves high quantification accuracy across different faults, with the I-V reconstruction error below 3%, confirming the feasibility and effectiveness of the application of differentiable physical simulators for PV system fault diagnosis.

cs.LG↗

arXiv2025

IF-Bench: Benchmarking and Enhancing MLLMs for Infrared Images with Generative Visual Prompting

Recent advances in multimodal large language models (MLLMs) have led to impressive progress across various benchmarks. However, their capability in understanding infrared images remains unexplored. To address this gap, we introduce IF-Bench, the first high-quality benchmark designed for evaluating multimodal understanding of infrared images. IF-Bench consists of 499 images sourced from 23 infrared datasets and 680 carefully curated visual question-answer pairs, covering 10 essential dimensions of image understanding. Based on this benchmark, we systematically evaluate over 40 open-source and closed-source MLLMs, employing cyclic evaluation, bilingual assessment, and hybrid judgment strategies to enhance the reliability of the results. Our analysis reveals how model scale, architecture, and inference paradigms affect infrared image comprehension, providing valuable insights for this area. Furthermore, we propose a training-free generative visual prompting (GenViP) method, which leverages advanced image editing models to translate infrared images into semantically and spatially aligned RGB counterparts, thereby mitigating domain distribution shifts. Extensive experiments demonstrate that our method consistently yields significant performance improvements across a wide range of MLLMs. The benchmark and code are available at https://github.com/casiatao/IF-Bench.

cs.CV↗

arXiv2025

Unveiling non-Hermitian band structures with non-Bloch supercells

Real-valued band structures are foundational to analyzing periodic systems within the Hermitian description and have been experimentally well-established over recent decades. In contrast, non-Hermitian systems exhibit complex band structures where both energy and momentum have imaginary parts, underpinning phenomena like the non-Hermitian skin effect and anomalous bulk-boundary correspondence that defy conventional Bloch theory. Experimentally mapping these complex bands-relating complex momentum to complex energy-and identifying their associated eigenstates is crucial for understanding these systems but remains a significant challenge. Here, we introduce a non-Bloch supercell framework designed to overcome this challenge by decoupling Bloch phase control from the imaginary part of momentum. Our method combines an exponent-flattening protocol with twisted boundary conditions, enabling system-size-independent control of imaginary momentum while preserving high-resolution Bloch phase sampling. Implemented in programmable one- and two-dimensional acoustic crystals, our approach acquires momentum-resolved complex energy surfaces and biorthogonal eigenmodes by Green's function measurements.Data obtained using this framework accurately predict open-boundary spectra and eigenstates, findings we verify through separate open-geometry experiments. Our work provides a broadly applicable experimental toolkit for exploring non-Hermitian band geometry and topology in diverse engineered classical and quantum platforms.

quant-ph↗

arXiv2025

Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering

Knowledge-based visual question answering (KB-VQA) requires visual language models (VLMs) to integrate visual understanding with external knowledge retrieval. Although retrieval-augmented generation (RAG) achieves significant advances in this task by combining knowledge-base querying, it still struggles with the quality of multimodal queries and the relevance of retrieved results. To overcome these challenges, we propose a novel three-stage method, termed Wiki-PRF, including Processing, Retrieval and Filtering stages. The processing stage dynamically invokes visual tools to extract precise multimodal information for retrieval. The retrieval stage integrates visual and text features to achieve multimodal knowledge retrieval. The filtering stage performs relevance filtering and concentration on retrieval results. To this end, we introduce a visual language model trained with answer accuracy and format consistency as reward signals via a reinforcement learning manner. This enhances the model's reasoning, tool invocation for accurate queries, and filtering of irrelevant content. Experiments on benchmark datasets (E-VQA and InfoSeek) show significant improvements~(36.0 and 42.8) in answer quality, achieving state-of-the-art performance. Code is available at https://github.com/cqu-student/Wiki-PRF

cs.CV↗

arXiv2025

Diffusion Model as a Noise-Aware Latent Reward Model for Step-Level Preference Optimization

Preference optimization for diffusion models aims to align them with human preferences for images. Previous methods typically use Vision-Language Models (VLMs) as pixel-level reward models to approximate human preferences. However, when used for step-level preference optimization, these models face challenges in handling noisy images of different timesteps and require complex transformations into pixel space. In this work, we show that pre-trained diffusion models are naturally suited for step-level reward modeling in the noisy latent space, as they are explicitly designed to process latent images at various noise levels. Accordingly, we propose the Latent Reward Model (LRM), which repurposes components of the diffusion model to predict preferences of latent images at arbitrary timesteps. Building on LRM, we introduce Latent Preference Optimization (LPO), a step-level preference optimization method conducted directly in the noisy latent space. Experimental results indicate that LPO significantly improves the model's alignment with general, aesthetic, and text-image alignment preferences, while achieving a 2.5-28x training speedup over existing preference optimization methods. Our code and models are available at https://github.com/Kwai-Kolors/LPO.

cs.CV↗

arXiv2025

Stabilization and Destabilization of Multimode Solitons in Nonlinear Degenerate Multi-Pass Cavities

Optical solitons in multimode nonlinear optical systems offer a unique platform for exploring the interplay of nonlinearity, dispersion, and spatial mode coupling, offering insights into complex nonlinear wave phenomena. Multi-pass cavities (MPCs) incorporating nonlinear Kerr media serve as prototypical systems, enabling high-efficiency supercontinuum generation and pulse compression. However, stabilizing femtosecond laser pulses in solid-medium-based MPCs (solid MPCs) under strong Kerr nonlinearity remains a significant challenge due to multimode coupling, which disrupts beam stability. In this work, we address this challenge by investigating the stability of laser pulses in MPCs using Floquet and perturbation model. We identify novel mode-coupling-suppression (MCS) medium lengths, where destructive interference among multimode wave components suppresses coupling and facilitates soliton stabilization. Under MCS conditions, our simulations demonstrate stable beam propagation in solid MPCs with nonlinear phases up to 1.5π per pass, achieving >13-fold pulse compression with excellent spatio-spectral homogeneity. Our findings offer valuable guidance for designing advanced MPCs with tailored Kerr media.

physics.optics↗

arXiv2025

Soliton Transitions Mediated by Skin-Mode Localization and Band Nonreciprocity

Solitons, typically resulting from a competition between band dispersion and nonlinearity, occur in lattices featuring the non-Hermitian skin effect as nonlinearity increases, accompanied by a transition in localization from linear skin modes to solitons. However, localization does not disentangle the role of skin modes in the soliton formation from that of band dispersion. Here, in such lattices, we uncover two distinct soliton phases, skin-mode-assisted solitons (SMASs) and nonreciprocity-dressed solitons (NRDSs). Rooted in fundamentally different mechanisms, SMASs originate from skin effect, while NRDSs stem from band nonreciprocity, each exhibiting unique spatial profiles. Using a stacked Su-Schrieffer-Heeger-like model as a prototype, we delineate the phase diagram of SMASs and NRDSs, each having clear phase boundaries. To interpret them, we formulate a Wannier-function-based nonlinear Hamiltonian, showing that soliton formation depends critically on how skin-mode localization and band nonreciprocity suppress or enhance wave dispersion. For SMASs, skin-mode localization reduces wave broadening at the localization sites, thereby lowering the formation threshold. This soliton phase is observable from edge dynamics and accompanied by a dynamical stability reentrance when transitioning from linear skin modes. In contrast, NRDSs, as well as their thresholds, originate from bulk band nonreciprocity and persist under periodic boundary conditions. Our framework offers predictive tools for characterizing and engineering solitons in experimentally realizable non-Hermitian systems, spanning optics to mechanics.

nlin.PS↗

arXiv2025

EvoVLMA: Evolutionary Vision-Language Model Adaptation

Pre-trained Vision-Language Models (VLMs) have been exploited in various Computer Vision tasks (e.g., few-shot recognition) via model adaptation, such as prompt tuning and adapters. However, existing adaptation methods are designed by human experts, requiring significant time cost and experience. Inspired by recent advances in Large Language Models (LLMs) based code generation, we propose an Evolutionary Vision-Language Model Adaptation (EvoVLMA) method to automatically search training-free efficient adaptation algorithms for VLMs. We recognize feature selection and logits computation as the key functions in training-free VLM adaptation, and propose a two-stage LLM-assisted evolutionary algorithm for optimizing these parts in a sequential manner, effectively addressing the challenge posed by the expansive search space through a divide-and-conquer strategy. Besides, to enhance the stability and efficiency of searching process, we propose low-precision code conversion, web based code execution and process monitoring, leading to a highly effective automatic algorithm design system. Extensive experiments demonstrate that the algorithms found by EvoVLMA can obtain promising results compared to previous manually-designed ones. More specifically, in the 8-shot image classification setting, the classical APE algorithm can be improved by 1.91 points in recognition accuracy. This research opens new possibilities for automating the optimization of adaptation algorithms of pre-trained multimodal models. Code is available at: https://github.com/kding1225/EvoVLMA

cs.CV↗

arXiv2025

Complex Frequency Fingerprint: Basic Concept and Theory

We introduce the complex frequency fingerprint (CFF), an experimentally accessible method for detecting the complex frequency Green's function (GF). Unlike the real frequency GF, where $ω$ is real, this complex frequency GF is shown to play a necessary role in both non-Hermitian and quantum many-body systems. For non-Hermitian systems, we will prove that our method detects complex energy spectra, eigenstates, and complex frequency GFs throughout the complex plane, providing necessary identification of the non-Hermitian skin effect. For quantum many-body systems, our method reveals quasiparticle peaks across the complex plane and intuitively illustrates interaction effects. This information is difficult to obtain with real frequency detection. Our method paves the way for exploring exotic phenomena in both non-Hermitian and quantum many-body systems, bridging theory and experiment across diverse physical areas.

cond-mat.mes-hall↗

arXiv2025

Re-ranking Reasoning Context with Tree Search Makes Large Vision-Language Models Stronger

Recent advancements in Large Vision Language Models (LVLMs) have significantly improved performance in Visual Question Answering (VQA) tasks through multimodal Retrieval-Augmented Generation (RAG). However, existing methods still face challenges, such as the scarcity of knowledge with reasoning examples and erratic responses from retrieved knowledge. To address these issues, in this study, we propose a multimodal RAG framework, termed RCTS, which enhances LVLMs by constructing a Reasoning Context-enriched knowledge base and a Tree Search re-ranking method. Specifically, we introduce a self-consistent evaluation mechanism to enrich the knowledge base with intrinsic reasoning patterns. We further propose a Monte Carlo Tree Search with Heuristic Rewards (MCTS-HR) to prioritize the most relevant examples. This ensures that LVLMs can leverage high-quality contextual reasoning for better and more consistent responses. Extensive experiments demonstrate that our framework achieves state-of-the-art performance on multiple VQA datasets, significantly outperforming In-Context Learning (ICL) and Vanilla-RAG methods. It highlights the effectiveness of our knowledge base and re-ranking method in improving LVLMs. Our code is available at https://github.com/yannqi/RCTS-RAG.

cs.CV↗

arXiv2025

Nanoscale Casimir Force Softening Originated from Quantum Surface Responses

Strong coupling between vacuum fields and quantum matter occurs at the nanoscale and broadens the horizon of light-matter interaction. Nanoscale Casimir force, as an exhibition of vacuum fields, inevitably experiences the influence of surface electrons due to their quantum character, which are ignorable in micron Casimir force. Here, we develop a three-dimensional conformal map method to tackle typical experimental configurations with surface electron contributions to Casimir force purposely and delicately included. Based on this method, we reveal that surface electrons can either enhance or suppress the nanoscale Casimir force, depending on materials and crystal facets. The mechanism is demonstrated to be the Casimir force softening, which results from surface electrons effectively altering the distance seen by the Casimir interaction. Our findings not only highlight the interaction between surface electrons and vacuum fields but also provide a recipe for theoretical and experimental investigation of nanoscale fluctuation-type problems.

cond-mat.mes-hall↗

arXiv2025

UNIP: Rethinking Pre-trained Attention Patterns for Infrared Semantic Segmentation

Pre-training techniques significantly enhance the performance of semantic segmentation tasks with limited training data. However, the efficacy under a large domain gap between pre-training (e.g. RGB) and fine-tuning (e.g. infrared) remains underexplored. In this study, we first benchmark the infrared semantic segmentation performance of various pre-training methods and reveal several phenomena distinct from the RGB domain. Next, our layerwise analysis of pre-trained attention maps uncovers that: (1) There are three typical attention patterns (local, hybrid, and global); (2) Pre-training tasks notably influence the pattern distribution across layers; (3) The hybrid pattern is crucial for semantic segmentation as it attends to both nearby and foreground elements; (4) The texture bias impedes model generalization in infrared tasks. Building on these insights, we propose UNIP, a UNified Infrared Pre-training framework, to enhance the pre-trained model performance. This framework uses the hybrid-attention distillation NMI-HAD as the pre-training target, a large-scale mixed dataset InfMix for pre-training, and a last-layer feature pyramid network LL-FPN for fine-tuning. Experimental results show that UNIP outperforms various pre-training methods by up to 13.5\% in average mIoU on three infrared segmentation tasks, evaluated using fine-tuning and linear probing metrics. UNIP-S achieves performance on par with MAE-L while requiring only 1/10 of the computational cost. Furthermore, UNIP significantly surpasses state-of-the-art (SOTA) infrared or RGB segmentation methods and demonstrates broad potential for application in other modalities, such as RGB and depth. Our code is available at https://github.com/casiatao/UNIP.

cs.CV↗

arXiv2025

Algebraic skin effect in two-dimensional non-Hermitian metamaterials

Metamaterials have unlocked unprecedented control over light by leveraging novel mechanisms to expand their functionality. Non-Hermitian physics further enhances the tunability of non-Hermitian metamaterials (NHMs) through phenomena such as the non-Hermitian skin effect (NHSE), enabling applications like directional amplification. The higher-dimensional NHSE manifests unique effects, including the algebraic skin effect (ASE), which features power-law decay instead of exponential localization, allowing for quasi-long-range interactions. In this work, we establish apparent criteria for achieving ASE in two-dimensional reciprocal NHMs with anisotropic and complex dielectric tensors. By numerically and theoretically demonstrating ASE through mismatched optical axes and geometric structures, we reveal that ASE is governed by a generalized Fermi surface whose dimensionality exceeds that of the Fermi surface. We further propose and validate a realistic photonic crystal design for ASE, which is experimentally accessible. Our recipe for ASE provides a versatile pathway for broader generalizations, including three-dimensional structures, synthetic dimensions, and other classical wave systems, paving the way for advancements in non-Hermitian photonics.

physics.optics↗

arXiv2025

Experimentally Probing Non-Hermitian Spectral Transition and Eigenstate Skewness

Non-Hermitian (NH) systems exhibit intricate spectral topology arising from complex-valued eigenenergies, with positive/negative imaginary parts representing gain/loss. Unlike the orthogonal eigenstates of Hermitian systems, NH systems feature left and right eigenstates that form a biorthogonal basis and can differ significantly, showcasing pronounced skewness between them. These characteristics give rise to unique properties absent in Hermitian systems, such as the NH skin effect and ultra spectral sensitivity. However, conventional experimental techniques are inadequate for directly measuring the complex-valued spectra and left and right eigenstates -- key elements for enhancing our knowledge of NH physics. This challenge is particularly acute in higher-dimensional NH systems, where the spectra and eigenstates are highly sensitive to macroscopic shapes, lattice geometry, and boundary conditions, posing greater experimental demands compared to one-dimensional systems. Here, we present a Green's function-based method that enables the direct measurement and characterization of both complex-valued energy spectra and the left and right eigenstates in arbitrary NH lattices. Using active acoustic crystals as the experimental platform, we observe spectral transitions and eigenstate skewness in two-dimensional NH lattices under both nonreciprocal and reciprocal conditions, with varied geometries and boundary conditions. Our approach renders complex spectral topology and left eigenstates experimentally accessible and practically meaningful, providing new insights into these quantities. The results not only confirm recent theoretical predictions of higher-dimensional NH systems but also establish a universal and versatile framework for investigating complex spectral properties and NH dynamics across a wide range of physical platforms.

cond-mat.mes-hall↗