arXiv subjects

Ling Ma

Publications and source records attributed to Ling Ma.

9 recordsLinked to original sources

arXiv2026

OmniCAD: A Large-Scale Benchmark for 3D Spatial Reasoning in Robotics Assemblies

Recent vision-language models (VLMs) show strong capabilities in robotic perception and spatial reasoning, yet their ability to reason about complex mechanical assemblies remains underexplored. We introduce OmniCAD, a large-scale benchmark for assembly-aware 3D spatial reasoning across diverse industrial systems, including robotic mechanisms, automotive components, aerospace structures, and agricultural machinery. OmniCAD contains 25k mechanical assemblies, with an average of 12 parts per assembly and 21 types of mate relationships. Each assembly includes a human-verified ground-truth 3D model and renderings from 20 viewpoints. The benchmark evaluates three capabilities: (1) component-level 3D spatial reasoning, requiring prediction of part positions and orientations; (2) part-to-part relational reasoning, requiring identification of mating relationships and assembly constraints; and (3) tool-augmented agentic reasoning, where models iteratively select viewpoints, inspect visual evidence, and refine predictions. Experiments show that current VLMs struggle with industrial assembly reasoning, often producing inaccurate poses, invalid mating relationships, part interpenetration, and degraded performance as assembly complexity increases. We will open-source the benchmark, evaluation code, and tool interfaces to support research on accurate, physically valid, and scalable 3D assembly reasoning.

cs.CV↗

arXiv2026

OmniMech: All-in-one Multimodal Mechanical Benchmark for 3D Reconstruction

Recent vision-language models (VLMs) can generate executable CAD programs from images, but existing methods mainly target coarse, general-purpose 3D objects and rarely address the fine-grained geometry and millimeter-level tolerances required in industrial mechanical design. We introduce OmniMech, the first million-scale benchmark for evaluating VLMs on executable CAD generation from industrial manufacturing data. OmniMech contains more than 251,000 fully dimensioned and toleranced 2D orthographic drawings, paired with native CAD models, multi-view renderings, mesh, STEP and B-rep representations, and rich semantic annotations. The benchmark includes four tasks: (1) parametric CAD program synthesis from engineering drawings; (2) diagram-to-3D reasoning for geometrically and structurally consistent reconstruction; (3) annotation-grounded reasoning over dimensions, symbols, feature callouts, and manufacturing constraints; and (4) tool-augmented agentic reasoning using visualization, measurement, CAD execution, and verification tools. Experiments show that current VLMs and CAD-specialized models still struggle with executable program synthesis, fine-grained 3D reconstruction, and reliable enforcement of dimensions and tolerances. We will release the benchmark data, evaluation code, and tool interfaces to support future research.

cs.CV↗

arXiv2026

OmniRouting: A Semantic-Coupled Multimodal Benchmark for Constraint-Aware Spatial Reasoning in PCB Routing

Recent large language models (LLMs) have demonstrated remarkable progress in constraint-aware navigation, maze reasoning, and graph reasoning. However, their ability to reason about complex routing problems under strict geometric, topological, and electrical constraints remains largely unexplored, despite routing being one of the most challenging and critical stages of electronic design automation (EDA). To bridge this gap, we introduce OmniRouting, the first large-scale benchmark designed to evaluate LLMs on printed-circuit-board (PCB) routing reasoning under real-world industrial design-rule, manufacturability, and connectivity constraints. OmniRouting contains 1,681 industrial-grade schematic-coupled PCB designs, including board geometries, routable component placements by human engineers, footprints, pad locations, netlists, stackup information, and routing constraints. The benchmark comprises four tasks: (1) geometric routing reasoning, generating physically valid copper traces, vias, and layer assignments to connect circuit nets within constrained board regions; (2) design-rule-aware routing reasoning, producing routable layouts that satisfy clearance, trace-width, via, obstacle-avoidance, and board-boundary constraints; (3) electrical functionality reasoning, preserving schematic-specified connectivity while reasoning over net names and functional roles to produce electrically correct routing; and (4) tool-augmented agentic routing, leveraging external tools for tasks (1)-(3). Our results reveal substantial limitations of current LMMs in PCB routing, including weak path-planning capabilities, poor adherence to design-rule constraints, and inconsistent preservation of electrical functionality. We will open-source all benchmark data, evaluation code, and tool interfaces to facilitate future research.

cs.CV↗

arXiv2024

Integral action feedback design for conservative abstract systems in the presence of input nonlinearities

In this article, we present a stabilization feedback law with integral action for conservative abstract linear systems subjected to actuator nonlinearity. Based on the designed control law, we first prove the well-posedness and global asymptotic stability of the origin of the closed-loop system by constructing a weak Lyapunov functional. Secondly, as an illustration, we apply the results to a wave equation coupled with an ordinary differential equation (ODE) at the boundary. Finally, we give the simulation results to illustrate the effectiveness of our method.

math.OC↗

arXiv2021

Joint modeling of geometric features of longitudinal process and discrete survival time measured on nested timescales: an application to fecundity studies

In biomedical studies, longitudinal processes are collected till time-to-event, sometimes on nested timescales (example, days within months). Most of the literature in joint modeling of longitudinal and time-to-event data has focused on modeling the mean or dispersion of the longitudinal process with the hazard for time-to-event. However, based on the motivating studies, it may be of interest to investigate how the cycle-level {\it geometric features} (such as the curvature, location and height of a peak), of a cyclical longitudinal process is associated with the time-to-event being studied. We propose a shared parameter joint model for a cyclical longitudinal process and a discrete survival time, measured on nested timescales, where the cycle-varying geometric feature is modeled through a linear mixed effects model and a proportional hazards model for the discrete survival time. The proposed approach allows for prediction of survival probabilities for future subjects based on their available longitudinal measurements. Our proposed model and approach is illustrated through simulation and analysis of Stress and Time-to-Pregnancy, a component of Oxford Conception Study. A joint modeling approach was used to assess whether the cycle-specific geometric features of the lutenizing hormone measurements, such as its peak or its curvature, are associated with time-to-pregnancy (TTP).

stat.ME↗

arXiv2021

AMA-GCN: Adaptive Multi-layer Aggregation Graph Convolutional Network for Disease Prediction

Recently, Graph Convolutional Networks (GCNs) have proven to be a powerful mean for Computer Aided Diagnosis (CADx). This approach requires building a population graph to aggregate structural information, where the graph adjacency matrix represents the relationship between nodes. Until now, this adjacency matrix is usually defined manually based on phenotypic information. In this paper, we propose an encoder that automatically selects the appropriate phenotypic measures according to their spatial distribution, and uses the text similarity awareness mechanism to calculate the edge weights between nodes. The encoder can automatically construct the population graph using phenotypic measures which have a positive impact on the final results, and further realizes the fusion of multimodal information. In addition, a novel graph convolution network architecture using multi-layer aggregation mechanism is proposed. The structure can obtain deep structure information while suppressing over-smooth, and increase the similarity between the same type of nodes. Experimental results on two databases show that our method can significantly improve the diagnostic accuracy for Autism spectrum disorder and breast cancer, indicating its universality in leveraging multimodal data for disease prediction.

cs.AI↗

arXiv2020

Contraction Mapping of Feature Norms for Classifier Learning on the Data with Different Quality

The popular softmax loss and its recent extensions have achieved great success in the deep learning-based image classification. However, the data for training image classifiers usually has different quality. Ignoring such problem, the correct classification of low quality data is hard to be solved. In this paper, we discover the positive correlation between the feature norm of an image and its quality through careful experiments on various applications and various deep neural networks. Based on this finding, we propose a contraction mapping function to compress the range of feature norms of training images according to their quality and embed this contraction mapping function into softmax loss or its extensions to produce novel learning objectives. The experiments on various classification applications, including handwritten digit recognition, lung nodule classification, face verification and face recognition, demonstrate that the proposed approach is promising to effectively deal with the problem of learning on the data with different quality and leads to the significant and stable improvements in the classification accuracy.

cs.CV↗

arXiv2019

Optimization of drug controlled release from multi-laminated devices based on the modified Tikhonov regularization method

From the viewpoint of inverse problem, the optimization of drug release based on the multi-laminated drug controlled release devices has been regarded as the solution problem of the diffusion equation initial value inverse problem. In view of the ill-posedness of the corresponding inverse problem, a modified Tikhonov regularization method is proposed by constructing a new regularizing filter function based on the singular value theory of compact operator. The convergence and the optimal asymptotic order of the regularized solution are obtained. Then the classical Tikhonov regularization method and the modified Tikhonov regularization method are applied to the optimization problem of the initial drug concentration distribution. For three various desired release profiles (constant release, linear decrease release and linear increase followed by a constant release profiles), better results can be obtained by using the modified Tikhonov regularization method. The numerical results demonstrate that the modified Tikhonov regularization method not only has the optimal asymptotic order, but also is suitable for the optimization and design of multi-laminated drug controlled release devices.

math.NA↗

arXiv2018

Rapid 3D Reconstruction of Indoor Environments to Generate Virtual Reality Serious Games Scenarios

Virtual Reality (VR) for Serious Games (SGs) is attracting increasing attention for training applications due to its potential to provide significantly enhanced learning to users. Some examples of the application of VR for SGs are complex training evacuation problems such as indoor earthquake evacuation or fire evacuation. The indoor 3D geometry of existing buildings can largely influence evacuees' behaviour, being instrumental in the design of VR SGs storylines and simulation scenarios. The VR scenarios of existing buildings can be generated from drawings and models. However, these data may not reflect the 'as-is' state of the indoor environment and may not be suitable to reflect dynamic changes of the system (e.g. Earthquakes), resulting in excessive development efforts to design credible and meaningful user experience. This paper explores several workflows for the rapid and effective reconstruction of 3D indoor environments of existing buildings that are suitable for earthquake simulations. These workflows start from Building Information Modelling (BIM), laser scanning and 360-degree panoramas. We evaluated the feasibility and efficiency of different approaches by using an earthquake-based case study developed for VR SGs.

cs.HC↗