arXiv subjects

Pan Du

Publications and source records attributed to Pan Du.

At least 19 recordsLinked to original sources

arXiv2026

DiFVM: A Vectorized Graph-Based Finite Volume Solver for Differentiable CFD on Unstructured Meshes

Differentiable programming has emerged as a structural prerequisite for gradient-based inverse problems and end-to-end hybrid physics--machine learning in computational fluid dynamics. However, existing differentiable CFD platforms are confined to structured Cartesian grids, excluding the geometrically complex domains where body-conforming unstructured discretizations are indispensable. We present DiFVM, the first GPU-accelerated, end-to-end differentiable finite-volume CFD solver operating natively on unstructured polyhedral meshes. The key enabling insight is a structural isomorphism between finite-volume discretization and graph neural network message-passing: by reformulating all FVM operators as static scatter/gather primitives on the mesh connectivity graph, DiFVM transforms irregular unstructured connectivity into a first-class GPU data structure. All operations are implemented in JAX/XLA, providing just-in-time compilation, operator fusion, and automatic differentiation through the complete simulation pipeline. Differentiable Windkessel outlet boundary conditions are provided for cardiovascular applications, and DiFVM accepts standard OpenFOAM case directories without modification for seamless adoption in existing workflows. Forward validation across benchmarks spanning canonical flows to patient-specific hemodynamics demonstrates close agreement with OpenFOAM, and end-to-end differentiability is demonstrated through inference of Windkessel parameters from sparse observations. DiFVM bridges the critical gap between differentiable programming and unstructured-mesh CFD, enabling gradient-based inverse problems and physics-integrated machine learning on complex engineering geometries.

cs.MS↗

arXiv2026

K-Sort Eval: Efficient Preference Evaluation for Visual Generation via Corrected VLM-as-a-Judge

The rapid development of visual generative models raises the need for more scalable and human-aligned evaluation methods. While the crowdsourced Arena platforms offer human preference assessments by collecting human votes, they are costly and time-consuming, inherently limiting their scalability. Leveraging vision-language model (VLMs) as substitutes for manual judgments presents a promising solution. However, the inherent hallucinations and biases of VLMs hinder alignment with human preferences, thus compromising evaluation reliability. Additionally, the static evaluation approach lead to low efficiency. In this paper, we propose K-Sort Eval, a reliable and efficient VLM-based evaluation framework that integrates posterior correction and dynamic matching. Specifically, we curate a high-quality dataset from thousands of human votes in K-Sort Arena, with each instance containing the outputs and rankings of K models. When evaluating a new model, it undergoes (K+1)-wise free-for-all comparisons with existing models, and the VLM provide the rankings. To enhance alignment and reliability, we propose a posterior correction method, which adaptively corrects the posterior probability in Bayesian updating based on the consistency between the VLM prediction and human supervision. Moreover, we propose a dynamic matching strategy, which balances uncertainty and diversity to maximize the expected benefit of each comparison, thus ensuring more efficient evaluation. Extensive experiments show that K-Sort Eval delivers evaluation results consistent with K-Sort Arena, typically requiring fewer than 90 model runs, demonstrating both its efficiency and reliability.

cs.CV↗

arXiv2025

Conditional neural field for spatial dimension reduction of turbulence data: a comparison study

We investigate conditional neural fields (CNFs), mesh-agnostic, coordinate-based decoders conditioned on a low-dimensional latent, for spatial dimensionality reduction of turbulent flows. CNFs are benchmarked against Proper Orthogonal Decomposition and a convolutional autoencoder within a unified encoding-decoding framework and a common evaluation protocol that explicitly separates in-range (interpolative) from out-of-range (strict extrapolative) testing beyond the training horizon, with identical preprocessing, metrics, and fixed splits across all baselines. We examine three conditioning mechanisms: (i) activation-only modulation (often termed FiLM), (ii) low-rank weight and bias modulation (termed FP), and (iii) last-layer inner-product coupling, and introduce a novel domain-decomposed CNF that localizes complexities. Across representative turbulence datasets (WMLES channel inflow, DNS channel inflow, and wall pressure fluctuations over turbulent boundary layers), CNF-FP achieves the lowest training and in-range testing errors, while CNF-FiLM generalizes best for out-of-range scenarios once moderate latent capacity is available. Domain decomposition significantly improves out-of-range accuracy, especially for the more demanding datasets. The study provides a rigorous, physics-aware basis for selecting conditioning, capacity, and domain decomposition when using CNFs for turbulence compression and reconstruction.

physics.flu-dyn↗

arXiv2025

Generative Latent Diffusion Model for Inverse Modeling and Uncertainty Analysis in Geological Carbon Sequestration

Geological Carbon Sequestration (GCS) has emerged as a promising strategy for mitigating global warming, yet its effectiveness heavily depends on accurately characterizing subsurface flow dynamics. The inherent geological uncertainty, stemming from limited observations and reservoir heterogeneity, poses significant challenges to predictive modeling. Existing methods for inverse modeling and uncertainty quantification are computationally intensive and lack generalizability, restricting their practical utility. Here, we introduce a Conditional Neural Field Latent Diffusion (CoNFiLD-geo) model, a generative framework for efficient and uncertainty-aware forward and inverse modeling of GCS processes. CoNFiLD-geo synergistically combines conditional neural field encoding with Bayesian conditional latent-space diffusion models, enabling zero-shot conditional generation of geomodels and reservoir responses across complex geometries and grid structures. The model is pretrained unconditionally in a self-supervised manner, followed by a Bayesian posterior sampling process, allowing for data assimilation for unseen/unobserved states without task-specific retraining. Comprehensive validation across synthetic and real-world GCS scenarios demonstrates CoNFiLD-geo's superior efficiency, generalization, scalability, and robustness. By enabling effective data assimilation, uncertainty quantification, and reliable forward modeling, CoNFiLD-geo significantly advances intelligent decision-making in geo-energy systems, supporting the transition toward a sustainable, net-zero carbon future.

physics.geo-ph↗

arXiv2025

AortaDiff: Volume-Guided Conditional Diffusion Models for Multi-Branch Aortic Surface Generation

Accurate 3D aortic construction is crucial for clinical diagnosis, preoperative planning, and computational fluid dynamics (CFD) simulations, as it enables the estimation of critical hemodynamic parameters such as blood flow velocity, pressure distribution, and wall shear stress. Existing construction methods often rely on large annotated training datasets and extensive manual intervention. While the resulting meshes can serve for visualization purposes, they struggle to produce geometrically consistent, well-constructed surfaces suitable for downstream CFD analysis. To address these challenges, we introduce AortaDiff, a diffusion-based framework that generates smooth aortic surfaces directly from CT/MRI volumes. AortaDiff first employs a volume-guided conditional diffusion model (CDM) to iteratively generate aortic centerlines conditioned on volumetric medical images. Each centerline point is then automatically used as a prompt to extract the corresponding vessel contour, ensuring accurate boundary delineation. Finally, the extracted contours are fitted into a smooth 3D surface, yielding a continuous, CFD-compatible mesh representation. AortaDiff offers distinct advantages over existing methods, including an end-to-end workflow, minimal dependency on large labeled datasets, and the ability to generate CFD-compatible aorta meshes with high geometric fidelity. Experimental results demonstrate that AortaDiff performs effectively even with limited training data, successfully constructing both normal and pathologically altered aorta meshes, including cases with aneurysms or coarctation. This capability enables the generation of high-quality visualizations and positions AortaDiff as a practical solution for cardiovascular research.

cs.CV↗

arXiv2025

HUG-VAS: A Hierarchical NURBS-Based Generative Model for Aortic Geometry Synthesis and Controllable Editing

Accurate, patient-specific vascular geometry is pivotal for diagnosis, planning, and device design, yet existing statistical shape modeling (SSM) pipelines rely on linear priors and topology-specific preprocessing that limit realism, scalability, and interoperability. We present HUG-VAS, a Hierarchical NURBS Generative framework for Vascular models, that unifies NURBS-based 3D shape encoding with diffusion-based generative modeling to synthesize fine-grained, CFD-ready aortic anatomies. HUG-VAS factorizes shape into (i) vessel centerlines generated by a denoising diffusion model and (ii) cross-sectional radius profiles synthesized by a classifier-free guided diffusion model conditioned on the centerline, thereby decoupling and preserving stochastic variability across these two anatomical layers. Beyond unconditional synthesis, we enable training-free, zero-shot conditional generation via diffusion posterior sampling from image-derived prompts (e.g., sparse 3D points, slice contours, or partial surface patches), supporting interactive semi-automatic segmentation, editing and robust reconstruction under degraded imaging. Trained on 21 patient-specific MRA cases, HUG-VAS generates multi-branch aortas with supra-aortic vessels whose biomarker distributions closely match the source cohort, and whose watertight NURBS outputs directly integrate with downstream CFD solvers. To our knowledge, this is the first SSM framework that bridges image-derived priors and generative shape synthesis through a unified combination of NURBS parameterization, hierarchical diffusion, and DPS, enabling a practical path from limited clinical anatomic information to simulation-ready vascular geometry.

cs.CV↗

arXiv2025

Batched Self-Consistency Improves LLM Relevance Assessment and Ranking

LLM query-passage relevance assessment is typically studied using a one-by-one pointwise (PW) strategy where each LLM call judges one passage at a time. However, this strategy requires as many LLM calls as there are passages while also preventing information sharing between passages. We thus hypothesize that batched PW methods, which evaluate multiple passages per LLM call, can improve not only efficiency but also judgment quality -- by enabling content from multiple passages to be seen jointly. Moreover, batched PW methods may be better suited to harness the test-time scaling benefits of self-consistency -- the ensembling technique of repeating (potentially perturbed) LLM tasks in parallel and aggregating results -- since batching can naturally enable prompt diversification through varied batch permutations and compositions to create more robust ensembles. We evaluate several batched PW methods against one-by-one PW and listwise ranking baselines on LLM relevance assessment and ranking tasks, using three passage retrieval datasets and GPT-4o, Claude Sonnet 3, and Amazon Nova Pro. We show that batching can greatly amplify self-consistency benefits, making batched PW methods achieve the best performance while often reducing latency by an order of magnitude or more compared to one-by-one PW methods. For instance, on legal search, batched PW ranking with GPT-4o improves from 43.8% to 51.3% NDCG@10 when using 1 vs. 15 self-consistency calls, compared to one-by-one PW ranking improving from 44.9% to 46.8% and being 15.3x slower.

cs.IR↗

arXiv2025

Unsupervised Learning for Class Distribution Mismatch

Class distribution mismatch (CDM) refers to the discrepancy between class distributions in training data and target tasks. Previous methods address this by designing classifiers to categorize classes known during training, while grouping unknown or new classes into an "other" category. However, they focus on semi-supervised scenarios and heavily rely on labeled data, limiting their applicability and performance. To address this, we propose Unsupervised Learning for Class Distribution Mismatch (UCDM), which constructs positive-negative pairs from unlabeled data for classifier training. Our approach randomly samples images and uses a diffusion model to add or erase semantic classes, synthesizing diverse training pairs. Additionally, we introduce a confidence-based labeling mechanism that iteratively assigns pseudo-labels to valuable real-world data and incorporates them into the training process. Extensive experiments on three datasets demonstrate UCDM's superiority over previous semi-supervised methods. Specifically, with a 60% mismatch proportion on Tiny-ImageNet dataset, our approach, without relying on labeled data, surpasses OpenMatch (with 40 labels per class) by 35.1%, 63.7%, and 72.5% in classifying known, unknown, and new classes.

cs.CV↗

arXiv2025

Implicit Neural Differential Model for Spatiotemporal Dynamics

Hybrid neural-physics modeling frameworks through differentiable programming have emerged as powerful tools in scientific machine learning, enabling the integration of known physics with data-driven learning to improve prediction accuracy and generalizability. However, most existing hybrid frameworks rely on explicit recurrent formulations, which suffer from numerical instability and error accumulation during long-horizon forecasting. In this work, we introduce Im-PiNDiff, a novel implicit physics-integrated neural differentiable solver for stable and accurate modeling of spatiotemporal dynamics. Inspired by deep equilibrium models, Im-PiNDiff advances the state using implicit fixed-point layers, enabling robust long-term simulation while remaining fully end-to-end differentiable. To enable scalable training, we introduce a hybrid gradient propagation strategy that integrates adjoint-state methods with reverse-mode automatic differentiation. This approach eliminates the need to store intermediate solver states and decouples memory complexity from the number of solver iterations, significantly reducing training overhead. We further incorporate checkpointing techniques to manage memory in long-horizon rollouts. Numerical experiments on various spatiotemporal PDE systems, including advection-diffusion processes, Burgers' dynamics, and multi-physics chemical vapor infiltration processes, demonstrate that Im-PiNDiff achieves superior predictive performance, enhanced numerical stability, and substantial reductions in memory and runtime cost relative to explicit and naive implicit baselines. This work provides a principled, efficient, and scalable framework for hybrid neural-physics modeling.

cs.LG↗

arXiv2025

JAX-BTE: A GPU-Accelerated Differentiable Solver for Phonon Boltzmann Transport Equations

This paper introduces JAX-BTE, a GPU-accelerated, differentiable solver for the phonon Boltzmann Transport Equation (BTE) based on differentiable programming. JAX-BTE enables accurate, efficient and differentiable multiscale thermal modeling by leveraging high-performance GPU computing and automatic differentiation. The solver efficiently addresses the high-dimensional and complex integro-differential nature of the phonon BTE, facilitating both forward simulations and data-augmented inverse simulations through end-to-end optimization. Validation is performed across a range of 1D to 3D simulations, including complex FinFET structures, in both forward and inverse settings, demonstrating excellent performance and reliability. JAX-BTE significantly outperforms state-of-the-art BTE solvers in forward simulations and uniquely enables inverse simulations, making it a powerful tool for multiscale thermal analysis and design for semiconductor devices.

physics.comp-ph↗

arXiv2025

AI-Powered Automated Model Construction for Patient-Specific CFD Simulations of Aortic Flows

Image-based modeling is essential for understanding cardiovascular hemodynamics and advancing the diagnosis and treatment of cardiovascular diseases. Constructing patient-specific vascular models remains labor-intensive, error-prone, and time-consuming, limiting their clinical applications. This study introduces a deep-learning framework that automates the creation of simulation-ready vascular models from medical images. The framework integrates a segmentation module for accurate voxel-based vessel delineation with a surface deformation module that performs anatomically consistent and unsupervised surface refinements guided by medical image data. By unifying voxel segmentation and surface deformation into a single cohesive pipeline, the framework addresses key limitations of existing methods, enhancing geometric accuracy and computational efficiency. Evaluated on publicly available datasets, the proposed approach demonstrates state-of-the-art performance in segmentation and mesh quality while significantly reducing manual effort and processing time. This work advances the scalability and reliability of image-based computational modeling, facilitating broader applications in clinical and research settings.

cs.CV↗

arXiv2025

Hierarchical LoG Bayesian Neural Network for Enhanced Aorta Segmentation

Accurate segmentation of the aorta and its associated arch branches is crucial for diagnosing aortic diseases. While deep learning techniques have significantly improved aorta segmentation, they remain challenging due to the intricate multiscale structure and the complexity of the surrounding tissues. This paper presents a novel approach for enhancing aorta segmentation using a Bayesian neural network-based hierarchical Laplacian of Gaussian (LoG) model. Our model consists of a 3D U-Net stream and a hierarchical LoG stream: the former provides an initial aorta segmentation, and the latter enhances blood vessel detection across varying scales by learning suitable LoG kernels, enabling self-adaptive handling of different parts of the aorta vessels with significant scale differences. We employ a Bayesian method to parameterize the LoG stream and provide confidence intervals for the segmentation results, ensuring robustness and reliability of the prediction for vascular medical image analysts. Experimental results show that our model can accurately segment main and supra-aortic vessels, yielding at least a 3% gain in the Dice coefficient over state-of-the-art methods across multiple volumes drawn from two aorta datasets, and can provide reliable confidence intervals for different parts of the aorta. The code is available at https://github.com/adlsn/LoGBNet.

cs.CV↗

arXiv2024

Personalized Clustering via Targeted Representation Learning

Clustering traditionally aims to reveal a natural grouping structure within unlabeled data. However, this structure may not always align with users' preferences. In this paper, we propose a personalized clustering method that explicitly performs targeted representation learning by interacting with users via modicum task information (e.g., $\textit{must-link}$ or $\textit{cannot-link}$ pairs) to guide the clustering direction. We query users with the most informative pairs, i.e., those pairs most hard to cluster and those most easy to miscluster, to facilitate the representation learning in terms of the clustering preference. Moreover, by exploiting attention mechanism, the targeted representation is learned and augmented. By leveraging the targeted representation and constrained contrastive loss as well, personalized clustering is obtained. Theoretically, we verify that the risk of personalized clustering is tightly bounded, guaranteeing that active queries to users do mitigate the clustering risk. Experimentally, extensive results show that our method performs well across different clustering tasks and datasets, even when only a limited number of queries are available.

cs.LG↗

arXiv2024

CoNFiLD-inlet: Synthetic Turbulence Inflow Using Generative Latent Diffusion Models with Neural Fields

Eddy-resolving turbulence simulations require stochastic inflow conditions that accurately replicate the complex, multi-scale structures of turbulence. Traditional recycling-based methods rely on computationally expensive precursor simulations, while existing synthetic inflow generators often fail to reproduce realistic coherent structures of turbulence. Recent advances in deep learning (DL) have opened new possibilities for inflow turbulence generation, yet many DL-based methods rely on deterministic, autoregressive frameworks prone to error accumulation, resulting in poor robustness for long-term predictions. In this work, we present CoNFiLD-inlet, a novel DL-based inflow turbulence generator that integrates diffusion models with a conditional neural field (CNF)-encoded latent space to produce realistic, stochastic inflow turbulence. By parameterizing inflow conditions using Reynolds numbers, CoNFiLD-inlet generalizes effectively across a wide range of Reynolds numbers ($Re_\tau$ between $10^3$ and $10^4$) without requiring retraining or parameter tuning. Comprehensive validation through a priori and a posteriori tests in Direct Numerical Simulation (DNS) and Wall-Modeled Large Eddy Simulation (WMLES) demonstrates its high fidelity, robustness, and scalability, positioning it as an efficient and versatile solution for inflow turbulence synthesis.

physics.flu-dyn↗

arXiv2024

CoNFiLD: Conditional Neural Field Latent Diffusion Model Generating Spatiotemporal Turbulence

This study introduces the Conditional Neural Field Latent Diffusion (CoNFiLD) model, a novel generative learning framework designed for rapid simulation of intricate spatiotemporal dynamics in chaotic and turbulent systems within three-dimensional irregular domains. Traditional eddy-resolved numerical simulations, despite offering detailed flow predictions, encounter significant limitations due to their extensive computational demands, restricting their applications in broader engineering contexts. In contrast, deep learning-based surrogate models promise efficient, data-driven solutions. However, their effectiveness is often compromised by a reliance on deterministic frameworks, which fall short in accurately capturing the chaotic and stochastic nature of turbulence. The CoNFiLD model addresses these challenges by synergistically integrating conditional neural field encoding with latent diffusion processes, enabling the memory-efficient and robust probabilistic generation of spatiotemporal turbulence under varied conditions. Leveraging Bayesian conditional sampling, the model can seamlessly adapt to a diverse range of turbulence generation scenarios without the necessity for retraining, covering applications from zero-shot full-field flow reconstruction using sparse sensor measurements to super-resolution generation and spatiotemporal flow data restoration. Comprehensive numerical experiments across a variety of inhomogeneous, anisotropic turbulent flows with irregular geometries have been conducted to evaluate the model's versatility and efficacy, showcasing its transformative potential in the domain of turbulence generation and the broader modeling of spatiotemporal dynamics.

physics.flu-dyn↗

arXiv2023

Semi-Supervised Learning via Weight-aware Distillation under Class Distribution Mismatch

Semi-Supervised Learning (SSL) under class distribution mismatch aims to tackle a challenging problem wherein unlabeled data contain lots of unknown categories unseen in the labeled ones. In such mismatch scenarios, traditional SSL suffers severe performance damage due to the harmful invasion of the instances with unknown categories into the target classifier. In this study, by strict mathematical reasoning, we reveal that the SSL error under class distribution mismatch is composed of pseudo-labeling error and invasion error, both of which jointly bound the SSL population risk. To alleviate the SSL error, we propose a robust SSL framework called Weight-Aware Distillation (WAD) that, by weights, selectively transfers knowledge beneficial to the target task from unsupervised contrastive representation to the target classifier. Specifically, WAD captures adaptive weights and high-quality pseudo labels to target instances by exploring point mutual information (PMI) in representation space to maximize the role of unlabeled data and filter unknown categories. Theoretically, we prove that WAD has a tight upper bound of population risk under class distribution mismatch. Experimentally, extensive results demonstrate that WAD outperforms five state-of-the-art SSL approaches and one standard baseline on two benchmark datasets, CIFAR10 and CIFAR100, and an artificial cross-dataset. The code is available at https://github.com/RUC-DWBI-ML/research/tree/main/WAD-master.

cs.CV↗

arXiv2022

PReGAN: Answer Oriented Passage Ranking with Weakly Supervised GAN

Beyond topical relevance, passage ranking for open-domain factoid question answering also requires a passage to contain an answer (answerability). While a few recent studies have incorporated some reading capability into a ranker to account for answerability, the ranker is still hindered by the noisy nature of the training data typically available in this area, which considers any passage containing an answer entity as a positive sample. However, the answer entity in a passage is not necessarily mentioned in relation with the given question. To address the problem, we propose an approach called \ttt{PReGAN} for Passage Reranking based on Generative Adversarial Neural networks, which incorporates a discriminator on answerability, in addition to a discriminator on topical relevance. The goal is to force the generator to rank higher a passage that is topically relevant and contains an answer. Experiments on five public datasets show that \ttt{PReGAN} can better rank appropriate passages, which in turn, boosts the effectiveness of QA systems, and outperforms the existing approaches without using external data.

cs.CL↗

arXiv2022

Deep learning-based surrogate model for 3-D patient-specific computational fluid dynamics

Optimization and uncertainty quantification have been playing an increasingly important role in computational hemodynamics. However, existing methods based on principled modeling and classic numerical techniques have faced significant challenges, particularly when it comes to complex 3D patient-specific shapes in the real world. First, it is notoriously challenging to parameterize the input space of arbitrarily complex 3-D geometries. Second, the process often involves massive forward simulations, which are extremely computationally demanding or even infeasible. We propose a novel deep learning surrogate modeling solution to address these challenges and enable rapid hemodynamic predictions. Specifically, a statistical generative model for 3-D patient-specific shapes is developed based on a small set of baseline patient-specific geometries. An unsupervised shape correspondence solution is used to enable geometric morphing and scalable shape synthesis statistically. Moreover, a simulation routine is developed for automatic data generation by automatic meshing, boundary setting, simulation, and post-processing. An efficient supervised learning solution is proposed to map the geometric inputs to the hemodynamics predictions in latent spaces. Numerical studies on aortic flows are conducted to demonstrate the effectiveness and merit of the proposed techniques.

physics.med-ph↗