arXiv subjects

Shan Gao

Publications and source records attributed to Shan Gao.

At least 19 recordsLinked to original sources

arXiv2026

Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding

Long-video understanding on edge devices must reason over hours of content under tight compute and bandwidth budgets. Subsampling visual tokens loses temporal structure, while text-only video memories lose fine-grained visual attributes. We observe a visual-textual duality: language memories carry long-range temporal structure better than dense frames, while pixels remain decisive for attribute-level perception. Building on this insight, we propose Caption-once, Frames-onDemand (CFD), a budget-aware edge-cloud agentic framework. The edge runs a single offline captioning pass that builds a dual-track narrative index, an event-level story skeleton plus a clip-level micro-log, cached and reused across queries without re-captioning. At query time, a cloud-side MLLM reasons over the index in a story-first loop centered on a lightweight Visual-Need Router: a per-query gating module that triggers bounded keyframe retrieval only for perceptual questions (appearance, on-screen text, attribute disambiguation) and keeps temporal-structural questions in language space. The router turns visual access into a first-class, query-conditioned cost, capping per-query frame consumption regardless of video length. Experiments on long-video benchmarks demonstrate strong accuracy-efficiency trade-offs while substantially reducing online visual processing.

cs.CV↗

arXiv2026

GrabVG: Graph-Attentive Binding for Visual Grounding in UAV Imagery

Visual grounding in Unmanned Aerial Vehicle (UAV) imagery aims to localize a target object in complex bird's-eye-view scenes according to a natural language description. However, the abundance of small, densely distributed, and visually similar objects creates high visual redundancy, while repetitive local configurations give rise to strong topological ambiguity. Existing approaches mainly focus on visual--language feature alignment or dense contextual interaction, yet they struggle to distinguish subtle inter-instance differences and effectively exploit spatial topological structures, leading to inaccurate grounding in highly crowded scenarios. To address these challenges, we propose $\textbf{GrabVG}$, a novel visual grounding framework inspired by human visual search. GrabVG explicitly decomposes grounding into two sequential stages: $\textit{preattentive hypothesis search}$ and $\textit{graph-attentive feature binding}$. Specifically, we first generate a compact set of reliable object hypotheses through distillation-guided proposal induction and text-aware hypothesis filtering, substantially reducing background distractions and semantic mismatches. These hypotheses are then organized into a sparse graph, where language-guided intra-instance visual cues and inter-instance topological relationships are jointly bound and propagated via graph attention, enabling efficient spatial reasoning and accurate target localization. Extensive experiments on AerialVG and AerialSense show that GrabVG achieves a favorable accuracy--speed trade-off, reaching 67.31$\%$ and 80.34$\%$ Acc@0.5 and outperforming the corresponding baselines by 10.55 and 8.76 percentage points, respectively.

cs.CV↗

arXiv2026

Surfsvr: 2D Surface Priors as 3D Geometric Regularizers for Sparse Voxel Reconstruction

Sparse voxel reconstruction offers an efficient representation for high-fidelity 3D modeling, yet its geometry is commonly optimized from local photometric evidence and discrete visibility statistics. This often leads to fragmented surfaces, excessive subdivision, and floating artifacts, particularly in weakly textured or sparsely observed regions. We introduce SurfSVR, a novel sparse voxel reconstruction paradigm that treats 2D surface priors as explicit 3D geometric regularizers. Instead of directly lifting noisy pixel-wise depth predictions, SurfSVR first organizes each image into coherent surface regions by jointly reasoning over appearance, monocular depth, normals and cross-view geometry. Each region is then represented by an adaptively selected planar or quadratic surface model based on fitting reliability and geometric complexity, while cross-model agreement distinguishes reliable geometry from ambiguous predictions. These structured 2D priors are lifted into 3D and integrated throughout the reconstruction pipeline. They guide surface-adaptive voxel subdivision, provide region-level depth and normal supervision during optimization, enhance geometrically reliable sparse-observed surfaces in voxel pruning, and suppress off-surface floaters during post-refinement training. This unified design converts semantic and geometric coherence in image space into persistent structural constraints in 3D. Extensive experiments on 3 public benchmarks demonstrate that SurfSVR consistently improves sparse voxel reconstruction across scenes with substantially different visibility and geometry characteristics, achieving state-of-the-art reconstruction quality. Codes and models will be released soon.

cs.CV↗

arXiv2026

From Scene-Centric to Observer-Centric: Modeling Observer-Aware Relations for 3D Scene Graph Generation

3D Scene Graph Generation (3DSGG) represents 3D scenes as structured object--relation--object graphs for spatial understanding. In observer-centric spatial perception, the same scene may be expressed under different local observer frames while its structure remains unchanged. However, existing models typically assume a fixed scene-aligned reference frame and may produce semantically inconsistent predictions when the scene is re-expressed in another observer frame. We attribute this failure to the heterogeneous frame dependency of relational predicates. Directional predicates such as $\textit{left}$, $\textit{front}$, $\textit{right}$, and $\textit{behind}$ are $\textbf{Observer-Dependent Relations}$, whereas most contact, support, and semantic predicates, such as $\textit{standing on}$ and $\textit{attached to}$, are approximately $\textbf{Observer-Independent Relations}$. Conventional models do not distinguish these frame responses, leading to degraded relation prediction under observer-frame reorientation. We introduce $\textbf{Observer-Aware Relations (OAR)}$, which combines observer-aware geometric encoding and relation specialization, supported by frame-stable object encoding, for unified multi-label predicate prediction. Experiments on 3DSSG show that OAR consistently outperforms baselines across controlled observer-frame reorientations without training-time frame-reorientation augmentation, while remaining competitive on the standard benchmark. The project page is available at https://oar-predicate.github.io/.

cs.CV↗

arXiv2026

Decoupling Intrinsic Molecular Efficacy from Platform Effects: An Interpretable Machine Learning Framework for Unbiased Perovskite Passivator Discovery

Rational design of interface passivators for perovskite solar cells is hindered by the entanglement of intrinsic molecular efficacy with extrinsic platform-dependent performance - a confounding factor that obscures true chemical advances. Here, we present a generalizable, interpretable machine learning framework that decouples these effects via an asymptotic saturation model, enabling unbiased discovery of molecules with genuine intrinsic gains. Trained on a curated dataset of 240 experimental entries, our model identifies hydrogen bond acceptor strength and electrostatic potential difference as key descriptors. Guided by these insights, we screened >121 million PubChem compounds using a hierarchical strategy integrating diversity clustering and uncertainty quantification. Five dual-functional candidates (e.g., TDZ-S, TZC-F) are identified, exhibiting superior predicted efficacy (surpassing experimental benchmarks) and high confidence. First-principles calculations confirm strong chemisorption (Eads<-1.7 eV), net electron donation, and optimized interfacial energetics. Crucially, our closed-loop "data-interpretation-screening-verification" pipeline establishes a transferable paradigm for rational materials design, extendable to other optoelectronic interfaces beyond perovskites.

cond-mat.mtrl-sci↗

arXiv2026

Generalized Aharonov-Bohm Effect

The Aharonov-Bohm (AB) effect highlights the fundamental role of electromagnetic potentials in quantum mechanics, manifesting as a phase shift for a charged particle in field-free regions. While well-established for static magnetic fluxes, the effect's behavior under time-varying fluxes remains an open and debated question. Employing the WKB method, we derive the AB phase shift for a time-dependent magnetic vector potential, demonstrating that for circular paths in the quasistatic regime, it is proportional to the time-averaged enclosed magnetic flux, $\Delta \phi_{\rm AB} = \frac{1}{T} \int_0^T e \Phi(t) \, dt$, with the total phase shift, including kinetic contributions, equaling $e \Phi(0)$. For non-circular paths, the phase shift depends on both the flux history and path geometry, revealing the effect's hybrid nature involving gauge potentials and induced electric fields. We verify the consistency of our gauge choice with Maxwell's equations and discuss the implications for local versus nonlocal interpretations of the AB effect. We also generalize the results to scenarios with nonzero external magnetic fields, where the enclosed flux is through the actual electron paths, and for circular paths of radius $R$, the AB phase shift is also proportional to the time average of the enclosed flux $\Phi_{\rm enc}(R,t)$, with the total phase shift depending only on the initial enclosed flux $e \Phi_{\rm enc}(R,0)$; for general non-circular paths, the external magnetic field affects trajectories and phase accumulation through the Lorentz force, leading to additional path dependence. These findings clarify the role of gauge-dependent potentials and induced fields in the generalized AB effect, offering new theoretical insights and potential applications in quantum technologies.

quant-ph↗

arXiv2026

From Joint to Single-System Psi-Onticity Without Preparation Independence

The Pusey-Barrett-Rudolph (PBR) theorem establishes $\psi$-onticity for individual quantum systems, but its standard formulation relies on the Preparation Independence Postulate (PIP). This has led to a prevalent view that rejecting PIP leaves open the possibility of $\psi$-epistemic models for individual systems. In this work, we show that this understanding is incomplete: once the PBR theorem establishes $\psi$-onticity for composite systems prepared in product states, the $\psi$-onticity of the individual subsystems follows directly from the tensor-product structure of quantum mechanics, without invoking PIP or any further auxiliary assumptions. This result removes a key auxiliary assumption from the PBR theorem, closes a persistent loophole for preserving $\psi$-epistemic models, and strengthens the conceptual foundations of $\psi$-ontology.

quant-ph↗

arXiv2026

Comment on "Aharonov-Bohm Phase is Locally Generated Like All Other Quantum Phases"

Marletto and Vedral [Phys. Rev. Lett. 125, 040401 (2020)] propose that the Aharonov-Bohm (AB) phase is locally mediated by entanglement between a charged particle and the quantized electromagnetic field, asserting gauge independence for non-closed paths. In this Comment, we critically analyze their model and demonstrate that the AB phase arises from the interaction with the vector potential $\mathbf{A}$, not from entanglement, which is a byproduct of the quantum electrodynamics (QED) framework. We show that their field-based energy formulation, intended to reflect local electromagnetic interactions, is mathematically flawed due to an incorrect prefactor and yields $ +q \mathbf{v} \cdot \mathbf{A}_{\mathbf{s}} $ in the Coulomb gauge, conflicting with QED's $ -q \mathbf{v} \cdot \mathbf{A}_{\mathbf{s}} $. This equivalence to $ q \mathbf{v} \cdot \mathbf{A}_{\mathbf{s}} $ holds only approximately in the Coulomb gauge under static conditions, failing for time-dependent fields and other gauges, undermining their claim of a gauge-independent local mechanism. Furthermore, we confirm that the AB phase is gauge-dependent for non-closed paths, contradicting their assertion. Our analysis reaffirms the conventional explanation in the semi-classical picture, where the AB phase is driven by the vector potential $\mathbf{A}$, with entanglement playing no causal role in its generation.

quant-ph↗

arXiv2026

Top 10 Open Challenges Steering the Future of Diffusion Language Model and Its Variants

The paradigm of Large Language Models (LLMs) is currently defined by auto-regressive (AR) architectures, which generate text through a sequential ``brick-by-brick'' process. Despite their success, AR models are inherently constrained by a causal bottleneck that limits global structural foresight and iterative refinement. Diffusion Language Models (DLMs) offer a transformative alternative, conceptualizing text generation as a holistic, bidirectional denoising process akin to a sculptor refining a masterpiece. However, the potential of DLMs remains largely untapped as they are frequently confined within AR-legacy infrastructures and optimization frameworks. In this Perspective, we identify ten fundamental challenges ranging from architectural inertia and gradient sparsity to the limitations of linear reasoning that prevent DLMs from reaching their ``GPT-4 moment''. We propose a strategic roadmap organized into four pillars: foundational infrastructure, algorithmic optimization, cognitive reasoning, and unified multimodal intelligence. By shifting toward a diffusion-native ecosystem characterized by multi-scale tokenization, active remasking, and latent thinking, we can move beyond the constraints of the causal horizon. We argue that this transition is essential for developing next-generation AI capable of complex structural reasoning, dynamic self-correction, and seamless multimodal integration.

cs.CL↗

arXiv2026

Towards Any-Quality Image Segmentation via Generative and Adaptive Latent Space Enhancement

Segment Anything Models (SAMs), known for their exceptional zero-shot segmentation performance, have garnered significant attention in the research community. Nevertheless, their performance drops significantly on severely degraded, low-quality images, limiting their effectiveness in real-world scenarios. To address this, we propose GleSAM++, which utilizes Generative Latent space Enhancement to boost robustness on low-quality images, thus enabling generalization across various image qualities. Additionally, to improve compatibility between the pre-trained diffusion model and the segmentation framework, we introduce two techniques, i.e., Feature Distribution Alignment (FDA) and Channel Replication and Expansion (CRE). However, the above components lack explicit guidance regarding the degree of degradation. The model is forced to implicitly fit a complex noise distribution that spans conditions from mild noise to severe artifacts, which substantially increases the learning burden and leads to suboptimal reconstructions. To address this issue, we further introduce a Degradation-aware Adaptive Enhancement (DAE) mechanism. The key principle of DAE is to decouple the reconstruction process for arbitrary-quality features into two stages: degradation-level prediction and degradation-aware reconstruction. Our method can be applied to pre-trained SAM and SAM2 with only minimal additional learnable parameters, allowing for efficient optimization. Extensive experiments demonstrate that GleSAM++ significantly improves segmentation robustness on complex degradations while maintaining generalization to clear images. Furthermore, GleSAM++ also performs well on unseen degradations, underscoring the versatility of our approach and dataset.

cs.CV↗

arXiv2026

Boosting Segment Anything Model to Generalize Visually Non-Salient Scenarios

Segment Anything Model (SAM), known for its remarkable zero-shot segmentation capabilities, has garnered significant attention in the community. Nevertheless, its performance is challenged when dealing with what we refer to as visually non-salient scenarios, where there is low contrast between the foreground and background. In these cases, existing methods often cannot capture accurate contours and fail to produce promising segmentation results. In this paper, we propose Visually Non-Salient SAM (VNS-SAM), aiming to enhance SAM's perception of visually non-salient scenarios while preserving its original zero-shot generalizability. We achieve this by effectively exploiting SAM's low-level features through two designs: Mask-Edge Token Interactive decoder and Non-Salient Feature Mining module. These designs help the SAM decoder gain a deeper understanding of non-salient characteristics with only marginal parameter increments and computational requirements. The additional parameters of VNS-SAM can be optimized within 4 hours, demonstrating its feasibility and practicality. In terms of data, we established VNS-SEG, a unified dataset for various VNS scenarios, with more than 35K images, in contrast to previous single-task adaptations. It is designed to make the model learn more robust VNS features and comprehensively benchmark the model's segmentation performance and generalizability on VNS scenarios. Extensive experiments across various VNS segmentation tasks demonstrate the superior performance of VNS-SAM, particularly under zero-shot settings, highlighting its potential for broad real-world applications. Codes and datasets are publicly available at https://guangqian-guo.github.io/VNS-SAM.

cs.CV↗

arXiv2025

Toward a benchmark for CTR prediction in online advertising: datasets, evaluation protocols and perspectives

This research designs a unified architecture of CTR prediction benchmark (Bench-CTR) platform that offers flexible interfaces with datasets and components of a wide range of CTR prediction models. Moreover, we construct a comprehensive system of evaluation protocols encompassing real-world and synthetic datasets, a taxonomy of metrics, standardized procedures and experimental guidelines for calibrating the performance of CTR prediction models. Furthermore, we implement the proposed benchmark platform and conduct a comparative study to evaluate a wide range of state-of-the-art models from traditional multivariate statistical to modern large language model (LLM)-based approaches on three public datasets and two synthetic datasets. Experimental results reveal that, (1) high-order models largely outperform low-order models, though such advantage varies in terms of metrics and on different datasets; (2) LLM-based models demonstrate a remarkable data efficiency, i.e., achieving the comparable performance to other models while using only 2% of the training data; (3) the performance of CTR prediction models has achieved significant improvements from 2015 to 2016, then reached a stage with slow progress, which is consistent across various datasets. This benchmark is expected to facilitate model development and evaluation and enhance practitioners' understanding of the underlying mechanisms of models in the area of CTR prediction. Code is available at https://github.com/NuriaNinja/Bench-CTR.

cs.IR↗

arXiv2025

WITNESS: A lightweight and practical approach to fine-grained predictive mutation testing

Existing fine-grained predictive mutation testing studies predominantly rely on deep learning, which faces two critical limitations in practice: (1) Exorbitant computational costs. The deep learning models adopted in these studies demand significant computational resources for training and inference acceleration. This introduces high costs and undermines the cost-reduction goal of predictive mutation testing. (2) Constrained applicability. Although modern mutation testing tools generate mutants both inside and outside methods, current fine-grained predictive mutation testing approaches handle only inside-method mutants. As a result, they cannot predict outside-method mutants, limiting their applicability in real-world scenarios. We propose WITNESS, a new fine-grained predictive mutation testing approach. WITNESS adopts a twofold design: (1) With collected features from both inside-method and outside-method mutants, WITNESS is suitable for all generated mutants. (2) Instead of using computationally expensive deep learning, WITNESS employs lightweight classical machine learning models for training and prediction. This makes it more cost-effective and enabling straightforward explanations of the decision-making processes behind the adopted models. Evaluations on Defects4J projects show that WITNESS consistently achieves state-of-the-art predictive performance across different scenarios. Additionally, WITNESS significantly enhances the efficiency of kill matrix prediction. Post-hoc analysis reveals that features incorporating information from before and after the mutation are the most important among those used in WITNESS. Test case prioritization based on the predicted kill matrix shows that WITNESS delivers results much closer to those obtained by using the actual kill matrix, outperforming baseline approaches.

cs.SE↗

arXiv2025

Identifying the Catalytic Descriptor of Single-Atom Catalysts in Nitrate Reduction Reaction: An Interpretable Machine-Learning Method

Elucidating the catalytic descriptor that accurately characterizes the structure-activity relationships of typical catalysts for various important heterogeneous catalytic reactions is pivotal for designing high-efficient catalytic systems. Here, an interpretable machine learning technique was employed to identify the key determinants governing the nitrate reduction reaction ($\rm NO_3RR$) performance across 286 single-atom catalysts (SACs) with the active sites anchored on double-vacancy $\rm BC_3$ monolayers. Through Shapley Additive Explanations (SHAP) analysis with reliable predictive accuracy, we quantitatively demonstrated that, favorable $\rm NO_3RR$ activity stems from a delicate balance among three critical factors: low $\rm N_V$, moderate $\rm D_N$, and specific doping patterns. Building upon these insights, we established a descriptor ($\psi$) that integrates the intrinsic catalytic properties and the intermediate O-N-H angle ($\theta$), effectively capturing the underlying structure-activity relationship. Guided by this, we further identified 16 promising catalysts with predicted low limiting potential ($U_{\rm L}$). Importantly, these catalysts are composed of cost-effective non-precious metal elements and are predicted to surpass most reported catalysts, with the best-performing Ti-V-1N1 is predicted to have an ultra-low $U_{\rm L}$ of $-0.10$ V.

physics.chem-ph↗

arXiv2025

Clean Code, Better Models: Enhancing LLM Performance with Smell-Cleaned Dataset

The Large Language Models (LLMs) have demonstrated great potential in code-related tasks. However, most research focuses on improving the output quality of LLMs (e.g., correctness), and less attention has been paid to the LLM input (e.g., the training code quality). Given that code smells are widely existed in practice and can negatively impact software maintainability and readability, this study takes the first systematic research to assess and improve dataset quality in terms of code smells. In this work, we first conduct a preliminary study to explore the presence of code smells in a popular benchmark dataset (i.e., CodeSearchNet-Python}) and evaluate the output of several popular LLMs (i.e., DeepSeek-Coder, CodeLlama, and MagiCoder), revealing that code smell issues extensively exist in LLM's input (e.g., benchmark dataset) and output (e.g., generated code). We then conduct our systematic research by taking three main steps: Firstly, we propose an LLM-based code smell cleaning tool, named SmellCC, which automatically refactors and removes code smells. To evaluate the correctness of the code refactoring, we construct a test set of 50 repositories sourced from the CodeSearchNet-Python benchmark for functional testing. Then we apply our curated smell-cleaned dataset to fine-tune two LLMs (i.e., DeepSeek-V2 and Qwen-Coder) to explore their potential for generating high-quality code. Thirdly, we investigate the impact of code smells on two downstream tasks: code completion and code search. Lastly, we derive several actionable implications for software engineering researchers and industry practitioners from our findings.

cs.SE↗

arXiv2025

ReLayout: Integrating Relation Reasoning for Content-aware Layout Generation with Multi-modal Large Language Models

Content-aware layout aims to arrange design elements appropriately on a given canvas to convey information effectively. Recently, the trend for this task has been to leverage large language models (LLMs) to generate layouts automatically, achieving remarkable performance. However, existing LLM-based methods fail to adequately interpret spatial relationships among visual themes and design elements, leading to structural and diverse problems in layout generation. To address this issue, we introduce ReLayout, a novel method that leverages relation-CoT to generate more reasonable and aesthetically coherent layouts by fundamentally originating from design concepts. Specifically, we enhance layout annotations by introducing explicit relation definitions, such as region, salient, and margin between elements, with the goal of decomposing the layout into smaller, structured, and recursive layouts, thereby enabling the generation of more structured layouts. Furthermore, based on these defined relationships, we introduce a layout prototype rebalance sampler, which defines layout prototype features across three dimensions and quantifies distinct layout styles. This sampler addresses uniformity issues in generation that arise from data bias in the prototype distribution balance process. Extensive experimental results verify that ReLayout outperforms baselines and can generate structural and diverse layouts that are more aligned with human aesthetics and more explainable.

cs.CV↗

arXiv2025

Comparing methods for handling missing data in electronic health records for dynamic risk prediction of central-line associated bloodstream infection

Electronic health records (EHR) often contain varying levels of missing data. This study compared different imputation strategies to identify the most suitable approach for predicting central line-associated bloodstream infection (CLABSI) in the presence of competing risks using EHR data. We analyzed 30862 catheter episodes at University Hospitals Leuven (2012-2013) to predict 7-day CLABSI risk using a landmark cause-specific supermodel, accounting for competing risks of hospital discharge and death. Imputation methods included simple methods (median/mode, last observation carried forward), multiple imputation, regression-based and mixed-effects models leveraging longitudinal structure, and random forest imputation to capture interactions and non-linearities. Missing indicators were also assessed alone and in combination with other imputation methods. Model performance was evaluated dynamically at daily landmarks up to 14 days post-catheter placement. The missing indicator approach showed the highest discriminative ability, achieving a mean AUROC of up to 0.782 and superior overall performance based on the scaled Brier score. Combining missing indicators with other methods slightly improved performance, with the mixed model approach combined with missing indicators achieving the highest AUROC (0.783) at day 4, and the missForestPredict approach combined with missing indicators yielding the best scaled Brier scores at earlier landmarks. This suggests that in EHR data, the presence or absence of information may hold valuable insights for patient risk prediction. However, the use of missing indicators requires caution, as shifts in EHR data over time can alter missing data patterns, potentially impacting model transportability.

stat.AP↗

arXiv2025

CIM-NET: A Video Denoising Deep Neural Network Model Optimized for Computing-in-Memory Architectures

While deep neural network (DNN)-based video denoising has demonstrated significant performance, deploying state-of-the-art models on edge devices remains challenging due to stringent real-time and energy efficiency requirements. Computing-in-Memory (CIM) chips offer a promising solution by integrating computation within memory cells, enabling rapid matrix-vector multiplication (MVM). However, existing DNN models are often designed without considering CIM architectural constraints, thus limiting their acceleration potential during inference. To address this, we propose a hardware-algorithm co-design framework incorporating two innovations: (1) a CIM-Aware Architecture, CIM-NET, optimized for large receptive field operation and CIM's crossbar-based MVM acceleration; and (2) a pseudo-convolutional operator, CIM-CONV, used within CIM-NET to integrate slide-based processing with fully connected transformations for high-quality feature extraction and reconstruction. This framework significantly reduces the number of MVM operations, improving inference speed on CIM chips while maintaining competitive performance. Experimental results indicate that, compared to the conventional lightweight model FastDVDnet, CIM-NET substantially reduces MVM operations with a slight decrease in denoising performance. With a stride value of 8, CIM-NET reduces MVM operations to 1/77th of the original, while maintaining competitive PSNR (35.11 dB vs. 35.56 dB

cs.CV↗