arXiv subjects

Ruizhi Wang

Publications and source records attributed to Ruizhi Wang.

17 recordsLinked to original sources

arXiv2026

Counterfactual Tests for Measuring Chain-of-Thought Faithfulness in Visual Language Models

Chain-of-thought (CoT) may often look plausible, yet it may not faithfully reflect the model's decision-making process. While methods for measuring the faithfulness of CoTs for textual inputs have been increasingly introduced, using these methods for visual inputs is not straightforward. In this work, we adapt the family of counterfactual methods for measuring CoT faithfulness, namely the Counterfactual Test (CT) and Correlational Counterfactual Test (CCT), to visual inputs, and call them vCT and vCCT, respectively. Using vCT and vCCT, we benchmark eight recent open-source Vision Language Models (VLMs) on two datasets. Our analysis shows that CoTs do not reliably track visual evidence that influences model predictions: they may omit the removed object even when its removal causes a large prediction shift, yet mention it when the shift is small. We further find that Predict-then-Explain explanations align more strongly with perturbation-induced probability shifts than pre-answer CoTs, while binary vCT scores are often nearly saturated. We also include a reconstruction control, in which images pass through the same editing pipeline without object removal, and find that the main object-removal intervention induces larger shifts than reconstruction alone. We construct and release Counter-SNLI-VE and Counter-A-OKVQA, two datasets of image pairs that differ by a single object.

cs.CV↗

arXiv2026

Large Language Models for Agentic NetOps and AIOps: Architectures, Evaluation, and Safety

Large language models (LLMs) are increasingly being used in network operations (NetOps) and artificial intelligence for IT operations (AIOps) for tasks ranging from telemetry retrieval and incident diagnosis to configuration planning and bounded remediation. As these systems acquire greater access to operational tools, the central question is no longer only what an LLM can do, but whether operational assurance increases commensurately with the authority granted to it. This survey examines that question through a structured, evidence-stratified review of agentic NetOps and AIOps. We organise the field around autonomy, tool scope, evidence traces, assurance controls, evaluation, security, and governance, and introduce an operational assurance contract that links each autonomy level to permitted tools, required evidence, independent gates, execution budgets, rollout and rollback duties, and audit requirements. The synthesis reveals a capability--assurance gap: evidence is comparatively strong for read-oriented assistance and tool-grounded diagnosis, but becomes substantially less complete as systems approach configuration change, bounded execution, and closed-loop operation. We therefore argue that evaluation should move beyond static question answering and model accuracy towards workflow-level assessment of evidence quality, tool use, policy and invariant compliance, staged execution, recovery, calibration, cost, and human intervention. We also examine prompt-borne attacks, poisoned or stale operational evidence, excessive agency, privilege boundaries, and weak auditability. Taken together, the survey frames agentic NetOps and AIOps as constrained operational control, in which useful autonomy depends on independently enforced assurance rather than model capability alone.

cs.NI↗

arXiv2026

ACQ: A Deployed Two-Stage Framework for Automated Creative Quota Allocation in Large-Scale Online Advertising

In digital advertising, demand-side platforms (DSPs) allow advertisers to create multiple ad creatives from a single photo for real-time bidding. While increasing the number of creatives can improve bidding opportunities, it cannot scale indefinitely, and the incremental advertising revenue typically exhibits diminishing returns as more creatives are generated. This raises a practical problem for DSPs: how to automatically determine an appropriate creative quota for each photo at scale. To address this problem, we propose Automated Creatives Quota (ACQ), a deployed two-stage framework for creative quota allocation in large-scale online advertising. In the first stage, ACQ predicts quota-conditioned expected revenue using a multi-task model built on an unbalanced binary tree, which is designed to handle the highly skewed revenue distribution across quota levels. In the second stage, ACQ formulates quota allocation under global capacity constraints as a multiple-choice knapsack problem (MCKP) and solves it with an efficient dual-based algorithm. Extensive offline experiments and online experiments on Kuaishou's advertising delivery platform demonstrate the effectiveness of ACQ, achieving a 6.20% increase in platform advertising revenue.

cs.AI↗

arXiv2026

Vision Hopfield Memory Networks for Image Recognition

Recent vision backbones, such as Transformer families and state-space models like Mamba, have achieved remarkable progress on image recognition. Despite their empirical success, these architectures remain far from the computational principles of the human brain, often demanding enormous amounts of training data while offering limited interpretability. We propose the Vision Hopfield Memory Network (V-HMN), a brain-inspired vision backbone that integrates hierarchical memory mechanisms across layers with iterative refinement updates. Specifically, V-HMN incorporates local Hopfield modules that provide associative memory dynamics at the image patch level, global Hopfield modules that function as episodic memory for contextual modulation, and a predictive-coding-inspired refinement rule for iterative error correction. By organizing these memory-based modules hierarchically, V-HMN captures both local and global dynamics in a unified framework. Memory retrieval exposes the relationship between inputs and stored patterns, providing a prototype-based form of interpretability through explicit memory retrieval, while the reuse of stored patterns improves data efficiency. This brain-inspired design therefore enhances data efficiency and provides a prototype-based form of interpretability compared to existing self-attention- or state-space-based approaches. We conducted extensive experiments on public image classification benchmarks. V-HMN achieves strong performance on small- and medium-scale benchmarks, and remains competitive with widely adopted backbone architectures on ImageNet despite minimal architectural tuning, while offering improved data efficiency and a prototype-based form of interpretability. These findings highlight the potential of V-HMN as a memory-centric alternative to standard vision backbones, thereby bridging brain-inspired computation with modern machine learning.

cs.LG↗

arXiv2026

Prototype Transformer: Towards Language Model Architectures Interpretable by Design

While state-of-the-art language models (LMs) surpass most humans in certain domains, their reasoning remains largely opaque, reducing trust and increasing the risk of deception and hallucination. We introduce the Prototype Transformer (ProtoT), an autoregressive LM architecture that replaces the quadratic-cost self-attention module of the Transformer with a linear-cost module based on prototypes, which are learned parameter vectors. In ProtoT, prototypes create communication channels that aggregate contextual information at different time scales. We show that this structure leads prototypes to automatically capture nameable concepts, such as "woman", during training, offering a path toward interpreting model reasoning and making targeted edits to model behavior. Compared with baselines, ProtoT scales well with model and data size, is robust to input perturbations, and performs well on text generation and downstream tasks, including GLUE. These results suggest that ProtoT is a promising step toward autoregressive language models that are more interpretable by design.

cs.AI↗

arXiv2026

A Stabilized Unfitted Space-time Finite Element Method for Parabolic Problems on Moving Domains

This paper presents a space-time finite element method (FEM) based on an unfitted mesh for solving parabolic problems on moving domains. Unlike other unfitted space-time finite element approaches that commonly employ the discontinuous Galerkin (DG) method for time-stepping, the proposed method employs a fully coupled space-time discretization. To stabilize the time-advection term, the streamline upwind Petrov-Galerkin (SUPG) scheme is applied in the temporal direction. A ghost penalty stabilization term is further incorporated to mitigate the small cut issue, thereby ensuring the well-conditioning of the stiffness matrix. Moreover, an a priori error estimate is derived in a discrete energy norm, which achieves an optimal convergence rate with respect to the mesh size. In particular, a space-time Poincare-Friedrichs inequality is established to support the condition number analysis. Several numerical examples are provided to validate the theoretical findings.

math.NA↗

arXiv2026

$D^3$-RSMDE: 40$\times$ Faster and High-Fidelity Remote Sensing Monocular Depth Estimation

Real-time, high-fidelity monocular depth estimation from remote sensing imagery is crucial for numerous applications, yet existing methods face a stark trade-off between accuracy and efficiency. Although using Vision Transformer (ViT) backbones for dense prediction is fast, they often exhibit poor perceptual quality. Conversely, diffusion models offer high fidelity but at a prohibitive computational cost. To overcome these limitations, we propose Depth Detail Diffusion for Remote Sensing Monocular Depth Estimation ($D^3$-RSMDE), an efficient framework designed to achieve an optimal balance between speed and quality. Our framework first leverages a ViT-based module to rapidly generate a high-quality preliminary depth map construction, which serves as a structural prior, effectively replacing the time-consuming initial structure generation stage of diffusion models. Based on this prior, we propose a Progressive Linear Blending Refinement (PLBR) strategy, which uses a lightweight U-Net to refine the details in only a few iterations. The entire refinement step operates efficiently in a compact latent space supported by a Variational Autoencoder (VAE). Extensive experiments demonstrate that $D^3$-RSMDE achieves a notable 11.85% reduction in the Learned Perceptual Image Patch Similarity (LPIPS) perceptual metric over leading models like Marigold, while also achieving over a 40x speedup in inference and maintaining VRAM usage comparable to lightweight ViT models.

cs.CV↗

arXiv2025

AMLP: Adjustable Masking Lesion Patches for Self-Supervised Medical Image Segmentation

Self-supervised masked image modeling (MIM) methods have shown promising performances on analyzing natural images. However, directly applying such methods to medical image segmentation tasks still cannot achieve satisfactory results. The challenges arise from the facts that (i) medical images are inherently more complex compared to natural images, and the subjects in medical images often exhibit more distinct contour features; (ii) moreover, the conventional high and fixed masking ratio in MIM is likely to mask the background, limiting the scope of learnable information. To address these problems, we propose a new self-supervised medical image segmentation framework, called Adjustable Masking Lesion Patches (AMLP), which employs Masked Patch Selection~(MPS) strategy to identify patches with high probabilities of containing lesions to help model achieve precise lesion reconstruction. To improve the categorization of patches in MPS, we further introduce Relative Reconstruction Loss (RRL) to better learn hard-to-reconstruct lesion patches. Then, Category Consistency Loss (CCL) is proposed to refine patch categorization based on reconstruction difficulty, enhancing difference between lesions and backgrounds. Moreover, an Adjustable Masking Ratio (AMR) strategy is proposed to gradually increase the masking ratio over training to expand~the scope of learnable mutual information. Extensive~experiments on two medical segmentation datasets demonstrate the superior performances of the proposed AMLP w.r.t. the SOTA self-supervised methods; the results prove that AMLP effectively addresses the challenges of applying masked modeling to medical images and capturing accurate lesion details that are crucial for segmentation tasks.

cs.CV↗

arXiv2025

RS3DBench: A Comprehensive Benchmark for 3D Spatial Perception in Remote Sensing

In this paper, we introduce a novel benchmark designed to propel the advancement of general-purpose, large-scale 3D vision models for remote sensing imagery. While several datasets have been proposed within the realm of remote sensing, many existing collections either lack comprehensive depth information or fail to establish precise alignment between depth data and remote sensing images. To address this deficiency, we present a visual Benchmark for 3D understanding of Remotely Sensed images, dubbed RS3DBench. This dataset encompasses 54,951 pairs of remote sensing images and pixel-level aligned depth maps, accompanied by corresponding textual descriptions, spanning a broad array of geographical contexts. It serves as a tool for training and assessing 3D visual perception models within remote sensing image spatial understanding tasks. Furthermore, we introduce a remotely sensed depth estimation model derived from stable diffusion, harnessing its multimodal fusion capabilities, thereby delivering state-of-the-art performance on our dataset. Our endeavor seeks to make a profound contribution to the evolution of 3D visual perception models and the advancement of geographic artificial intelligence within the remote sensing domain. The dataset, models and code will be accessed on the https://rs3dbench.github.io.

cs.CV↗

arXiv2025

Comprehensive characterization of nonlinear viscoelastic properties of arterial tissues using guided-wave optical coherence elastography

The mechanical properties of arterial walls are critical for maintaining vascular function under pulsatile pressure and are closely linked to the development of cardiovascular diseases. Despite advances in imaging and elastography, comprehensive characterization of the complex mechanical behavior of arterial tissues remains challenging. Here, we present a broadband guided-wave optical coherence elastography (OCE) technique, grounded in viscoelasto-acoustic theory, for quantifying the nonlinear viscoelastic, anisotropic, and layer-specific properties of arterial walls with high spatial and temporal resolution. Our results reveal a strong stretch dependence of arterial viscoelasticity, with increasing prestress leading to a reduction in tissue viscosity. Under mechanical loading, the adventitia becomes significantly stiffer than the media, attributable to engagement of collagen fibers. Chemical degradation of collagen fibers highlighted their role in nonlinear viscoelasticity. This study demonstrates the potential of OCE as a powerful tool for detailed profiling of vascular biomechanics, with applications in basic research and future clinical diagnosis.

physics.bio-ph↗

arXiv2023

C^2M-DoT: Cross-modal consistent multi-view medical report generation with domain transfer network

In clinical scenarios, multiple medical images with different views are usually generated simultaneously, and these images have high semantic consistency. However, most existing medical report generation methods only consider single-view data. The rich multi-view mutual information of medical images can help generate more accurate reports, however, the dependence of multi-view models on multi-view data in the inference stage severely limits their application in clinical practice. In addition, word-level optimization based on numbers ignores the semantics of reports and medical images, and the generated reports often cannot achieve good performance. Therefore, we propose a cross-modal consistent multi-view medical report generation with a domain transfer network (C^2M-DoT). Specifically, (i) a semantic-based multi-view contrastive learning medical report generation framework is adopted to utilize cross-view information to learn the semantic representation of lesions; (ii) a domain transfer network is further proposed to ensure that the multi-view report generation model can still achieve good inference performance under single-view input; (iii) meanwhile, optimization using a cross-modal consistency loss facilitates the generation of textual reports that are semantically consistent with medical images. Extensive experimental studies on two public benchmark datasets demonstrate that C^2M-DoT substantially outperforms state-of-the-art baselines in all metrics. Ablation studies also confirmed the validity and necessity of each component in C^2M-DoT.

cs.CV↗

arXiv2023

MvCo-DoT:Multi-View Contrastive Domain Transfer Network for Medical Report Generation

In clinical scenarios, multiple medical images with different views are usually generated at the same time, and they have high semantic consistency. However, the existing medical report generation methods cannot exploit the rich multi-view mutual information of medical images. Therefore, in this work, we propose the first multi-view medical report generation model, called MvCo-DoT. Specifically, MvCo-DoT first propose a multi-view contrastive learning (MvCo) strategy to help the deep reinforcement learning based model utilize the consistency of multi-view inputs for better model learning. Then, to close the performance gaps of using multi-view and single-view inputs, a domain transfer network is further proposed to ensure MvCo-DoT achieve almost the same performance as multi-view inputs using only single-view inputs.Extensive experiments on the IU X-Ray public dataset show that MvCo-DoT outperforms the SOTA medical report generation baselines in all metrics.

cs.CV↗

arXiv2023

MPS-AMS: Masked Patches Selection and Adaptive Masking Strategy Based Self-Supervised Medical Image Segmentation

Existing self-supervised learning methods based on contrastive learning and masked image modeling have demonstrated impressive performances. However, current masked image modeling methods are mainly utilized in natural images, and their applications in medical images are relatively lacking. Besides, their fixed high masking strategy limits the upper bound of conditional mutual information, and the gradient noise is considerable, making less the learned representation information. Motivated by these limitations, in this paper, we propose masked patches selection and adaptive masking strategy based self-supervised medical image segmentation method, named MPS-AMS. We leverage the masked patches selection strategy to choose masked patches with lesions to obtain more lesion representation information, and the adaptive masking strategy is utilized to help learn more mutual information and improve performance further. Extensive experiments on three public medical image segmentation datasets (BUSI, Hecktor, and Brats2018) show that our proposed method greatly outperforms the state-of-the-art self-supervised baselines.

cs.CV↗

arXiv2019

Non-equilibrium properties of graphene probed by superconducting tunnel spectroscopy

We report on non-equilibrium properties of graphene probed by superconducting tunnel spectroscopy. A hexagonal boron nitride (hBN) tunnel barrier in combination with a superconducting Pb contact is used to extract the local energy distribution function of the quasiparticles in graphene samples in different transport regimes. In the cases where the energy distribution function resembles a Fermi-Dirac distribution, the local electron temperature can directly be accessed. This allows us to study the cooling mechanisms of hot electrons in graphene. In the case of long samples (device length $L$ much larger than the electron-phonon scattering length $l_{e-ph}$), cooling through acoustic phonons is dominant. We find a cross-over from the dirty limit with a power law $~T^3$ at low temperature to the clean limit at higher temperatures with a power law $~T^4$ and a deformation potential of 13.3 eV. For shorter samples, where $L$ is smaller than $l_{e-ph}$ but larger than the electron-electron scattering length $l_{e-e}$, the well-known cooling through electron out-diffusion is found. Interestingly, we find strong indications of an enhanced Lorenz number in graphene. We also find evidence of a non-Fermi-Dirac distribution function, which is a result of non-interacting quasiparticles in very short samples.

cond-mat.mes-hall↗

arXiv2019

Spectrally-Resolved Photodynamics of Individual Emitters in Large-Area Monolayers of Hexagonal-Boron Nitride

Hexagonal boron nitride (h-BN) is a 2D, wide band-gap semiconductor that has recently been shown to display bright room-temperature emission in the visible region, sparking immense interest in the material for use in quantum applications. In this work, we study highly crystalline, single atomic layers of chemical vapour deposition (CVD)-grown hexagonal boron nitride and find predominantly one type of emissive state. Using a multidimensional super-resolution fluorescence microscopy technique we simultaneously measure spatial position, intensity and spectral properties of the emitters, as they are exposed to continuous wave illumination over minutes. As well as low emitter heterogeneity, we observe inhomogeneous broadening of emitter line-widths and power law dependency in fluorescence intermittency, this is in striking similarity to previous work on quantum dots. These results show that high control over h-BN growth and treatment can produce a narrow distribution of emitter type, and that surface interactions heavily influence the photodynamics. Furthermore, we highlight the utility of spectrally-resolved wide-field microscopy in the study of optically-active excitons in atomically thin two-dimensional materials.

physics.optics↗

arXiv2018

A peeling approach for integrated manufacturing of large mono-layer h-BN crystals

Hexagonal boron nitride (h-BN) is the only known material aside from graphite with a structure composed of simple, stable, non-corrugated atomically thin layers. While historically used as lubricant in powder form, h-BN layers have become particularly attractive as an ultimately thin insulator. Practically all emerging electronic and photonic device concepts rely on h-BN exfoliated from small bulk crystallites, which limits device dimensions and process scalability. Here, we address this integration challenge for mono-layer h-BN via a chemical vapour deposition process that enables crystal sizes exceeding 0.5 mm starting from commercial, reusable platinum foils, and in unison allows a delamination process for easy and clean layer transfer. We demonstrate sequential pick-up for the assembly of graphene/h-BN heterostructures with atomic layer precision, while minimizing interfacial contamination. Our process development builds on a systematic understanding of the underlying mechanisms. The approach can be readily combined with other layered materials and opens a scalable route to h-BN layer integration and reliable 2D material device layer stacks.

cond-mat.mtrl-sci↗

arXiv2018

Raman Spectral Indicators of Catalyst Decoupling for Transfer of CVD Grown 2D Materials

Through a combination of monitoring the Raman spectral characteristics of 2D materials grown on copper catalyst layers, and wafer scale automated detection of the fraction of transferred material, we reproducibly achieve transfers with over 97.5% monolayer hexagonal boron nitride and 99.7% monolayer graphene coverage, for up to 300 mm diameter wafers. We find a strong correlation between the transfer coverage obtained for graphene and the emergence of a lower wavenumber 2D- peak component, with the concurrent disappearance of the higher wavenumber 2D+ peak component during oxidation of the catalyst surface. The 2D peak characteristics can therefore act as an unambiguous predictor of the success of the transfer. The combined monitoring and transfer process presented here is highly scalable and amenable for roll-to-roll processing.

cond-mat.mtrl-sci↗