arXiv subjects

Qiang Zou

Publications and source records attributed to Qiang Zou.

At least 19 recordsLinked to original sources

arXiv2026

KDH-CAD: Knowledge-data hybrid CAD learning under data scarcity

Deep learning in computer-aided design (CAD) remains fundamentally constrained by the data scarcity challenge: authentic CAD data is difficult to collect at scale, while synthetic data may not faithfully reflect real design practice. Rather than pursuing ever-larger CAD datasets, this paper alternatively treats CAD learning as a knowledge completion and calibration problem. It introduces KDH-CAD, a knowledge-data hybrid framework that integrates pretrained knowledge in foundation models, structured domain knowledge from textbooks/tutorials, and a very small amount of labeled CAD data. Domain knowledge is used to elicit and complete CAD-relevant concepts that are weakly expressed or under-represented in pretrained foundation models, while labeled CAD data calibrates these concepts in the latent space to account for task-specific geometric variability, without fine-tuning the foundation model. Experiments on real-world mechanical part classification show that KDH-CAD achieves strong performance in low-data regimes, reaching 92.6\% accuracy with only 250 training samples, 95.8\% with 1,000 samples, and continuing to improve with additional data. This matches or exceeds state-of-the-art performance that typically requires an order of magnitude more data. These results suggest that combining pretrained foundation models with structured domain knowledge can substantially reduce reliance on large-scale CAD datasets, providing a principled and practical direction for data-efficient CAD learning.

cs.GR↗

arXiv2026

AirfoilGen: A valid-by-construction and performance-aware latent diffusion model for airfoil generation

Airfoil shape design is a fundamental task in aerospace engineering, with a direct impact on flight stability and fuel consumption. Deep learning has recently emerged as a promising tool for this task, but existing deep generative approaches remain limited in both geometric validity and physical controllability. They offer little control over the generated shapes, yielding invalid geometries, and they typically do not condition effectively on aerodynamic performance. To address these issues, this paper proposes AirfoilGen, a valid-by-construction and performance-aware latent diffusion model for airfoil. It first introduces a novel airfoil representation scheme, the circle sweeping representation, to constrain the generative process so that output shapes respect essential airfoil characteristics. It then enables explicit control over aerodynamic performance (e.g., lift and drag coefficients) by operating in a learned latent space: a transformer model encodes airfoil shapes into vector embeddings, and a conditional diffusion model denoises Gaussian noise into these latent embeddings while incorporating target aerodynamic performance. In addition, this paper presents a new dataset of over 200,000 airfoils, which is substantially larger than the widely used UIUC airfoil dataset (1,650 airfoils) and more suitable for training modern deep generative models. Experiments demonstrate that AirfoilGen enables airfoil generation with far greater geometric validity and aerodynamic performance controllability than previously achievable, with an average performance-conditioning accuracy of 98.41%.

cs.LG↗

arXiv2026

Generative and isoparametric geometric modeling of large-scale and multiscale microstructures

As additive manufacturing advances toward higher printing resolution and larger build volumes, microstructures can be designed with finer geometric features over larger physical domains. This trend poses a fundamental challenge for geometric modeling: massive geometric details must be represented compactly, while their associations across scales must be maintained consistently.Existing methods cannot scale well to this requirement. Explicit representations suffer from prohibitive memory cost, and implicit representations remain compact only when microstructures admit analytic, periodic, or otherwise concise procedural descriptions. This paper proposes a new geometric modeling method that treats microstructure modeling as an on-demand generative process, rather than requiring the full instantiation of all geometric details. We first develop ExVCC, an extended volumetric Catmull-Clark spline representation that enables local spline refinement to go beyond tensor-product topology. Built on ExVCC, we introduce new shape-coding schemes and refinement rules that compactly encode large-scale geometric details and enable their localized evaluation through on-demand hierarchical refinement. To model geometric details across scales, we further propose an isoparametric representation in which details across scales are defined over a shared parametric domain using the same family of spline bases of ExVCC. This formulation turns the ExVCC's spline refinement hierarchy into a common framework for geometry encoding, on-demand generation, and cross-scale association, allowing geometric modifications to propagate automatically across scales. The effectiveness of the proposed method is demonstrated through a series of examples and comparisons.

cs.GR↗

arXiv2026

Cubic Hermite Lattice Structures

Lattice structures are of growing importance in additive manufacturing, where complex internal geometries are increasingly required for lightweight, high surface-to-volume ratios, multifunctionality, and other superior mechanical properties. Conventional lattice modeling methods typically represent struts with simple primitives, such as cylinders or cones, limiting geometric diversity and the design space. Although recent efforts have increased strut-shape complexity to address this issue, they often do so at the expense of computational efficiency and modeling robustness. As a result, achieving both rich geometric expressiveness and efficient computation remains a challenging problem. In this paper, we present an implicit modeling method that expands the design and optimization space of lattice structures while preserving the modeling robustness and efficiency of implicit representations. In our method, each strut is defined as a convolution surface over a skeletal graph, and its profile shape is controlled by a cubic Hermite curve. By exploiting the polynomial structure of both the convolution kernel and the cubic Hermite curve-controlled profile, we derive analytical expressions for efficient field evaluation, avoiding costly and unstable numerical computation. Four case studies have been conducted to validate the proposed method in terms of profile shape diversity, graded lattice modeling, as well as slicing robustness and efficiency.

cs.GR↗

arXiv2025

Symmetry-required Orbital Selectivity in Monolayer FeSe

Orbital-selective correlations have been observed to play an important role in Fe-based superconductors. Here, in contrast to previous site-local Mott transition-based origins, we present a band-theory-based mechanism for orbital-selective physics in monolayer FeSe, for which only electron pockets appear. Underlying our mechanism is the observation in density functional theory (DFT) calculations that around the M point in the Brillouin zone, antiferromagnetic fluctuations are strongly coupled to electrons in $x^2-y^2$ orbitals but weakly coupled to those in $\{xz,yz\}$ orbitals. Symmetry-arguments reveal that this orbital selective coupling originates from the different intertwined orbital and Fe-site sublattice Bloch wavefunctions for these two sets of orbitals at the M point, specifically, the $x^2-y^2$ orbitals can be Fe-site localized. The strong coupling of electrons in $x^2-y^2$ orbitals to the magnetic fluctuations enables orbital-selective electronic renormalizations that can account for important features of our angle-resolved photoemission spectroscopy (ARPES) measurements. Our symmetry-required mechanism for orbital selective physics can be generalized to a range of crystal space groups with four-fold and six-fold screw axes.

cond-mat.str-el↗

arXiv2025

CADDesigner: Conceptual CAD Model Generation with a General-Purpose Agent

Computer-Aided Design (CAD) is widely used for conceptual design and parametric 3D modeling, but typically requires a high level of expertise from designers. To lower the entry barrier and facilitate early-stage CAD modeling, we present CADDesigner, an LLM-powered agent for conceptual CAD design. The agent accepts both textual descriptions and sketches as input, engaging in interactive dialogue with users to refine and clarify design requirements through comprehensive requirement analysis. Built upon a novel Explicit Context Imperative Paradigm (ECIP), the agent generates high-quality CAD modeling code. During the generation process, the agent incorporates iterative visual feedback to improve model quality. Generated design cases can be stored in a structured knowledge base, providing a mechanism for continual knowledge accumulation and future improvement of code generation. Experimental results show that CADDesigner achieves competitive performance and outperforms representative baselines on conceptual CAD model generation tasks.

cs.AI↗

arXiv2025

Correlation Enhanced Electron-Phonon Coupling in FeSe/SrTiO$_3$ at a Magic Angle

While a predictive theory for unconventional superconductivity in Fe-based superconductors remains elusive, an extensively debated aspect is the interaction between phonons and strongly correlated electrons, and its potential role in the pairing mechanism. Here, through the combination of first principles dynamical mean field theory calculations and epitaxial growth of the single-layer FeX (X=Se, S, Te) on SrTiO$_3$ (STO)(001) substrate, which facilitates the controlled distortion of the FeX$_4$ tetrahedron, we demonstrate an unique superconducting dome where the superconducting gap peaks at a `magic' angle of the FeX$_4$ tetrahedron and the electron-phonon coupling (EPC) for the A$_{1g}$ mode is maximized for the FeSe film. Our findings uncover a significant role of electronic correlations in strengthening Cooper pairing in unconventional superconductors by enhancing EPC.

cond-mat.str-el↗

arXiv2025

Light-driven lattice metastability for enhanced superconductivity in FeSe/SrTiO3

Driven quantum materials with on demand properties controlled by external stimuli are critical for emergent quantum technology. In optically tunable superconducting heterostructures, the lattice responses at the buried interface may hold the key to the light susceptibility but is very challenging to detect. In this work, a nondestructive synchrotron-based X-ray scattering phase-retrieval technique is implemented in monolayer-FeSe/SrTiO3 heterostructures to capture the three-dimensional interfacial atomic displacements in-situ as the interface superconductivity is actively manipulated by light. It is found that the interlayer sliding between FeSe and SrTiO3 can drastically alter how the lattice responds to the light. In domains with selected stacking configurations, the interface transforms the very weak photoexcitation in SrTiO3 into significant Fe-atom displacements in FeSe and generate metastable interfacial structures that can lead to a persistent superconductivity enhancement. These findings demonstrate an effective strategy for achieving greatly amplified light-lattice coupling for efficient quantum phase manipulations at designed interfaces.

cond-mat.mtrl-sci↗

arXiv2025

Matrix representation and GPU-optimized parallel B-spline computing

B-spline modeling is fundamental to CAD systems, and its evaluation and manipulation algorithms currently in use were developed decades ago, specifically for CPU architectures. While remaining effective for many applications, these algorithms become increasingly inadequate as CAD models grow more complex, such as large-scale assemblies and microstructures. GPU acceleration offers a promising solution, but most existing GPU B-spline algorithms simply adapt CPU counterparts without accounting for the mismatch between the unstructured, recursive nature of B-splines and the structured nature of GPU kernels, ultimately failing to fully leverage GPU capabilities. This paper presents a novel approach that transforms B-spline representations into regular matrix structures, reducing all evaluation and manipulation computations to matrix addition and multiplication, thus better aligning with GPU architecture. By combining this matrix representation with GPU-optimized task scheduling and memory access patterns, the paper demonstrates significant performance improvements in the key B-spline operations of inversion and projection. Experimental results show an improvement of about two orders of magnitude in computational speed compared to existing methods.

cs.DC↗

arXiv2025

Soap Film-inspired Subdivisional Lattice Structure Construction

Lattice structures, distinguished by their customizable geometries at the microscale and outstanding mechanical performance, have found widespread application across various industries. One fundamental process in their design and manufacturing is constructing boundary representation (B-rep) models, which are essential for running advanced applications like simulation, optimization, and process planning. However, this construction process presents significant challenges due to the high complexity of lattice structures, particularly in generating nodal shapes where robustness and smoothness issues can arise from the complex intersections between struts. To address these challenges, this paper proposes a novel approach for lattice structure construction by cutting struts and filling void regions with subdivisional nodal shapes. Inspired by soap films, the method generates smooth, shape-preserving control meshes using Laplacian fairing and subdivides them through the point-normal Loop (PN-Loop) subdivision scheme to obtain subdivisional nodal shapes. The proposed method ensures robust model construction with reduced shape deviations, enhanced surface fairness, and smooth transitions between subdivisional nodal shapes and retained struts. The effectiveness of the method has been demonstrated by a series of examples and comparisons. The code will be open-sourced upon publication.

cs.CG↗

arXiv2025

Bringing Attention to CAD: Boundary Representation Learning via Transformer

The recent rise of generative artificial intelligence (AI), powered by Transformer networks, has achieved remarkable success in natural language processing, computer vision, and graphics. However, the application of Transformers in computer-aided design (CAD), particularly for processing boundary representation (B-rep) models, remains largely unexplored. To bridge this gap, we propose a novel approach for adapting Transformers to B-rep learning, called the Boundary Representation Transformer (BRT). B-rep models pose unique challenges due to their irregular topology and continuous geometric definitions, which are fundamentally different from the structured and discrete data Transformers are designed for. To address this, BRT proposes a continuous geometric embedding method that encodes B-rep surfaces (trimmed and untrimmed) into Bezier triangles, preserving their shape and continuity without discretization. Additionally, BRT employs a topology-aware embedding method that organizes these geometric embeddings into a sequence of discrete tokens suitable for Transformers, capturing both geometric and topological characteristics within B-rep models. This enables the Transformer's attention mechanism to effectively learn shape patterns and contextual semantics of boundary elements in a B-rep model. Extensive experiments demonstrate that BRT achieves state-of-the-art performance in part classification and feature recognition tasks.

cs.GR↗

arXiv2025

Semantic Direct Modeling

Current direct modeling systems limit users to low-level interactions with vertices, edges, and faces, forcing designers to manage detailed geometric elements rather than focusing on high-level design intent. This paper introduces semantic direct modeling (SDM), a novel approach that lifts direct modeling from low-level geometric modifications to high-level semantic interactions. This is achieved by utilizing a large language model (LLM) fine-tuned with CAD-specific prompts, which can guide the LLM to reason through design intent and accurately interpret CAD commands, thereby allowing designers to express their intent using natural language. Additionally, SDM maps design intent to the corresponding geometric features in the CAD model through a new conditional, context-sensitive feature recognition method, which uses generative AI to dynamically assign feature labels based on design intent. Together, they enable a seamless flow from high-level design intent to low-level geometric modifications, bypassing tedious software interactions. The effectiveness of SDM has been validated through real mechanical design cases.

cs.HC↗

arXiv2025

PromptHash: Affinity-Prompted Collaborative Cross-Modal Learning for Adaptive Hashing Retrieval

Cross-modal hashing is a promising approach for efficient data retrieval and storage optimization. However, contemporary methods exhibit significant limitations in semantic preservation, contextual integrity, and information redundancy, which constrains retrieval efficacy. We present PromptHash, an innovative framework leveraging affinity prompt-aware collaborative learning for adaptive cross-modal hashing. We propose an end-to-end framework for affinity-prompted collaborative hashing, with the following fundamental technical contributions: (i) a text affinity prompt learning mechanism that preserves contextual information while maintaining parameter efficiency, (ii) an adaptive gated selection fusion architecture that synthesizes State Space Model with Transformer network for precise cross-modal feature integration, and (iii) a prompt affinity alignment strategy that bridges modal heterogeneity through hierarchical contrastive learning. To the best of our knowledge, this study presents the first investigation into affinity prompt awareness within collaborative cross-modal adaptive hash learning, establishing a paradigm for enhanced semantic consistency across modalities. Through comprehensive evaluation on three benchmark multi-label datasets, PromptHash demonstrates substantial performance improvements over existing approaches. Notably, on the NUS-WIDE dataset, our method achieves significant gains of 18.22% and 18.65% in image-to-text and text-to-image retrieval tasks, respectively. The code is publicly available at https://github.com/ShiShuMo/PromptHash.

cs.CV↗

arXiv2025

Unraveling Enhanced Superconductivity in Single-layer FeSe through Substrate Surface Terminations

Single-layer FeSe films grown on (001) SrTiO3 substrates have shown a significant increase in superconducting transition temperature compared to bulk FeSe. Several mechanisms have been proposed to explain such enhancement, including electron doping, interfacial electron-phonon coupling, and strong electron correlations. To pinpoint the primary driver, we grew FeSe films on SrTiO3 substrates with coexisting TiO2 and SrO surface terminations. Scanning tunneling spectroscopy revealed a larger superconducting gap of 17 meV for FeSe on TiO2 compared to 11 meV on SrO. Tunneling spectroscopy also showed a larger work function on SrO, leading to reduced charge transfer, as confirmed by angle-resolved photoemission spectroscopy. Scanning transmission electron microscopy revealed distinctive interfacial atomic-scale structures, with the Se-Fe-Se tetrahedral angle changing from 109.9{\deg} on SrO to 105.1{\deg} on TiO2. Compared to dynamical mean field theory calculations, these results suggest optimal electron correlations in FeSe/TiO2 for enhancing high-temperature superconductivity.

cond-mat.supr-con↗

arXiv2025

A Token-FCM based risk assessment method for complex engineering designs

Engineering design risks could cause unaffordable losses, and thus risk assessment plays a critical role in engineering design. On the other hand, the high complexity of modern engineering designs makes it difficult to assess risks effectively and accurately due to the complex two-way, dynamic causal-effect risk relations in engineering designs. To address this problem, this paper proposes a new risk assessment method called token fuzzy cognitive map (Token-FCM). Its basic idea is to model the two-way causal-risk relations with the FCM method, and then augment FCM with a token mechanism to model the dynamics in causal-effect risk relations. Furthermore, the fuzzy sets and the group decision-making method are introduced to initialize the Token-FCM method so that comprehensive and accurate risk assessments can be attained. The effectiveness of the proposed method has been demonstrated by a real example of engine design for a horizontal directional drilling machine.

cs.CE↗

arXiv2025

A Quasi-Optimal Shape Design Method for Lattice Structure Construction

Lattice structures, known for their superior mechanical properties, are widely used in industries such as aerospace, automotive, and biomedical. Their advantages primarily lie in the interconnected struts at the micro-scale. The robust construction of these struts is crucial for downstream design and manufacturing applications, as it provides a detailed shape description necessary for precise simulation and fabrication. However, constructing lattice structures presents significant challenges, particularly at nodes where multiple struts intersect. The complexity of these intersections can lead to robustness issues. To address this challenge, this paper presents an optimization-based approach that simplifies the construction of lattice structures by cutting struts and connecting them to optimized node shapes. By utilizing the recent Grey Wolf optimization method -- a type of meta-heuristic method -- for node shape design, the approach ensures robust model construction and optimal shape design. Its effectiveness has been validated through a series of case studies with increasing topological and geometric complexity.

cs.CE↗

arXiv2024

A review of geometric modeling methods in microstructure design and manufacturing

Microstructures, characterized by intricate structures at the microscopic scale, hold the promise of important disruptions in the field of mechanical engineering due to the superior mechanical properties they offer. One fundamental technique of microstructure design and manufacturing is geometric modeling, which generates the 3D computer models required to run high-level procedures such as simulation, optimization, and process planning. There is, however, a lack of comprehensive discussions on this body of knowledge. The goal of this paper is to compile existing microstructure modeling methods and clarify the challenges, progress, and limitations of current research. It also concludes with future research directions that may improve and/or complement current methods, such as compressive and generative microstructure representations. By doing so, the paper sheds light on what has already been made possible for microstructure modeling, what developments can be expected in the near future, and which topics remain problematic.

cs.CG↗

arXiv2024

Intelligent CAD 2.0

Integrating modern artificial intelligence (AI) techniques, particularly generative AI, holds the promise of revolutionizing computer-aided design (CAD) tools and the engineering design process. However, the direction of "AI+CAD" remains unclear: how will the current generation of intelligent CAD (ICAD) differ from its predecessor in the 1980s and 1990s, what strategic pathways should researchers and engineers pursue for its implementation, and what potential technical challenges might arise? As an attempt to address these questions, this paper investigates the transformative role of modern AI techniques in advancing CAD towards ICAD. It first analyzes the design process and reconsiders the roles AI techniques can assume in this process, highlighting how they can restructure the path humans, computers, and designs interact with each other. The primary conclusion is that ICAD systems should assume an intensional rather than extensional role in the design process. This offers insights into the evaluation of the previous generation of ICAD (ICAD 1.0) and outlines a prospective framework and trajectory for the next generation of ICAD (ICAD 2.0).

cs.HC↗