cs.CE — Pith

0

math.NA 2026-07-03

Space-time method keeps acoustics stable for long industrial runs

by Simon Schneider, Ceyhun Özdemir +3 more

A Stable Boundary Element Method for Reliable Long-Time Industrial Sound Emission

The boundary element scheme stays accurate over extended times where standard methods diverge, matching real measurements.

abstract click to expand

In this paper we investigate a stable space-time formulation for long-time industrial sound emission problems. To this end, we use a well-posed Galerkin formulation in space and time of the acoustic wave equation in $\mathbb{R}^3$, involving a hypersingular boundary integral operator. Our numerical experiments confirm that the resulting time stepping scheme is stable and accurate for complex acoustic problems in industrial geometries, in contrast to alternative well-known schemes. The proposed method is shown to be efficient for real-world problems, and we obtain very good agreement with physical acoustic measurements.

0

cs.CE 2026-07-03

DoubleML cuts error in retrofit energy-saving estimates

by Kevin Zalipski, David Zapata Gonzalez +1 more

Predicting Heterogeneous Treatment Effects Of Building Energy Saving Retrofits Using Causal Machine Learning

A simulation with known true effects shows orthogonalization removes bias from household traits that drive both adoption and consumption.

abstract click to expand

Information Systems research increasingly relies on machine learning (ML) to predict outcomes in complex sociotechnical systems, yet predictive models are not designed to identify causal effects. This limitation is particularly critical in building retrofits, where unbiased estimates of energy savings are essential for climate policy and investment decisions. Because retrofit adoption is shaped by household and building characteristics that also affect energy consumption, predictive ML can yield biased effect estimates. This paper systematically benchmarks leading causal ML estimators, including metalearners (S-, T- and X-Learners) and DoubleML across multiple retrofit interventions. To enable this comparison, we construct a physically grounded simulation in which true treatment effects and realistic adoption biases are known. Results show that DoubleML achieves the lowest estimation errors, particularly for complex envelope retrofits. These findings demonstrate that orthogonalising the treatment assignment improves causal effect estimation and provides a methodological foundation for large-scale energy retrofit and policy evaluation.

0

cs.CE 2026-07-03

LLM agent runs full topology optimization from text prompts

by Haoju Lin, Wenchang Zhang +4 more

TO-Master: an LLM-agent framework for automated topology optimization

TO-Master selects tools, builds models, and returns results for structural and thermal problems without any user code.

abstract click to expand

Topology optimization (TO) has become a mature computational design method, but using it still requires substantial manual effort in geometry preparation, mesh generation, boundary-condition assignment, solver setup, and postprocessing. This implementation barrier limits the use of TO outside expert workflows, even when differentiable finite element solvers are available. This work introduces TO-Master, a large language model (LLM) agent framework that turns finite-element-based TO into a conversational, tool-orchestrated workflow. From natural language instructions and optional mesh, geometry, or image inputs, the agent selects computational tools, constructs finite element TO models, checks meshes and boundary conditions, and launches sensitivity-based optimization with typed solver arguments. The framework supports generated and uploaded meshes, image-to-mesh conversion, 2D and 3D structural compliance minimization, thermal conduction, multiple load cases, stress-constrained optimization, and engineering geometries. Numerical experiments show that TO-Master can reproduce standard benchmark results and solve more complex engineering examples while returning optimized results, field distributions, convergence histories, and interactive artifacts without user-written code. An instruction ablation study further shows that tool-usage rules, internal reasoning guidance, and few-shot examples are critical for robust formulation under ambiguous user input. By combining LLM-agent orchestration with deterministic finite element and optimization tools, TO-Master removes the burden of trivial setup and routine model construction, lowers the modeling barrier of TO, and preserves a reliable numerical workflow. The TO-Master platform is available online at https://www.bohrium.com/en/apps/to-master.

0

eess.SY 2026-07-02

Region graph transfers mode shape recognition across vehicles

by Tong Duy Son, Marc Brughmans +5 more

Robust and Explainable 3D Mode Shape Recognition Using Region-Aware Graph Neural Networks

Mapping models to shared structural regions lets AI work on new designs without identical meshes or retraining on full data.

abstract click to expand

Mode shape recognition is a fundamental task in automotive NVH development, yet it remains dependent on manual visual inspection by experienced engineers. Existing approaches based on engineering heuristics, Modal Assurance Criterion (MAC), or geometry-dependent AI representations often exhibit limited robustness across different vehicle architectures, finite element (FE) meshes, and experimental measurement layouts, restricting their industrial applicability. This paper presents a Canonical Engineering Graph Representation and region-aware graph learning framework for robust and explainable 3D mode shape recognition. Rather than learning directly from vehicle-specific FE meshes, heterogeneous FE models and experimental measurements are transformed into a common graph whose nodes represent semantically meaningful structural regions connected through engineering-informed relationships. Geometry-independent regional descriptors are combined with graph attention learning and region-aware pooling to capture structural interactions while preserving engineering semantics and enabling physically interpretable predictions. The resulting representation decouples engineering knowledge from numerical discretization, allowing transfer across different vehicle programs without requiring identical mesh topology or sensor configurations. The proposed framework is validated using FE and experimental datasets from four vehicle programs under severe label scarcity. Results demonstrate high classification accuracy, cross-vehicle transferability, and physically meaningful explanations by directly relating predictions to engineering-defined structural regions used in NVH analysis. Beyond mode shape recognition, the proposed Canonical Engineering Graph Representation provides a reusable engineering abstraction for trustworthy and transferable AI across heterogeneous simulation and experimental workflows.

0

cs.CE 2026-07-02

FUSE links separate FEBio models for multiphysics without code changes

by Steve A. Maas, Farhan Muhib +1 more

FUSE: A Partitioned Field-Exchange Framework for Coupling Physics Simulations in FEBio

Time-decoupled field exchange reproduces coupled solutions for cartilage degradation and bone healing.

abstract click to expand

Computational biomechanics increasingly requires models that combine mechanics, transport, chemistry, and biological regulation across different spatial and temporal scales. The FEBio simulation software provides extensive open-source capabilities for modeling these processes using monolithic approaches. However, assembling independently developed physics models into reproducible coupled workflows remains challenging. Existing approaches often require custom scripts or external software pipelines, which can limit model reuse and complicate development. We present FUSE, the FEBio Unified Simulation and Exchange framework, a partitioned coupling plugin that enables separately defined FEBio models to communicate through structured field exchange. FUSE is designed for problems that are best solved independently, particularly when fast mechanical responses influence slower biological or chemical evolution. The framework uses a time-decoupled strategy in which a primary model advances on the longer time scale, while one or more secondary models are repeatedly initialized, supplied with updated fields, solved over shorter time horizons, with results returned to the primary model. Field exchange utilizes existing FEBio data maps, output fields, and user-specified filters, allowing coupled workflows to be constructed without modifying the underlying solvers. The framework was able to reproduce reference coupled solutions while handling bidirectional transfer, spatial field mapping, and filtered exchange of model variables. Example applications demonstrated coupling between mechanical loading and chemical degradation in injured cartilage and interaction between biological tissue formation and mechanical feedback during bone healing. By separating coupling logic from physics implementation, FUSE provides a practical mechanism for building maintainable multiphysics workflows within FEBio.

0

physics.comp-ph 2026-07-02

Sequential THM coupling matches analytical benchmarks

by J. Al Kubaisy, G. E. Hammond +5 more

Verification of a sequential thermo-poroelasticity formulation in PFLOTRAN

A non-iterative fixed-stress split solves flow and temperature first then mechanics, agreeing with solutions for pressure, temperature, and

abstract click to expand

We present the verification of a thermo--hydrologic--mechanical capability implemented within the PFLOTRAN framework, with emphasis on benchmark-based assessment of the THM implementation. The thermal--hydrologic (TH) equations for mass and energy balance are solved on control-volume blocks or Voronoi cells, while the quasi-static momentum balance is solved on an element-based dual mesh. The coupling is achieved using a strictly sequential, non-iterative fixed-stress split strategy in which the TH system is solved implicitly for pressure and temperature, followed by a mechanics update for the displacement unknowns. Several verification problems are set up against poroelastic and thermo-poroelastic benchmarks, demonstrating agreement with analytical or semi-analytical benchmark responses for pressure diffusion, the temperature field, and mechanical deformation. In addition, we propose a treatment for discontinuities (e.g., fractures) based on mapping between mechanical and flow degrees of freedom, and validate the approach by comparison to an analytical solution. This work establishes the basis for thermo-poroelastic coupling in PFLOTRAN and provides a solid modeling foundation for a range of applications (e.g., enhanced geothermal systems and other subsurface energy storage) involving coupled thermal--hydrologic--mechanical (THM) processes in geologic porous media.

0

cs.LG 2026-07-02

Multi-particle flow maps create Feynman-Kac corrector for global search

by Binglin Ji, Anindya Sarkar +3 more

Sequentially-Controlled Interactive Multi-Particle Flow-Maps for Online Feedback-Driven Search

Sample sharing across particles steers ensembles to KL-tilted targets while avoiding mode collapse in sequential alignment.

abstract click to expand

While generative models have enabled training-free reward alignment, current methods typically excel in local exploration within narrow regions of the underlying distribution. These approaches struggle when preferences are unknown a priori and only revealed through sequential feedback-a scenario demanding broad exploration to uncover high-utility regions. To address this, we propose Sequentially-Controlled Interactive Multi-Particle Flow-Maps (IMPFM), a framework for sample-efficient online feedback-driven search. IMPFM progressively transports a group of interactive particles toward the target distribution, maintaining the broad coverage essential for heterogeneous preference alignment. IMPFM introduces a principled and efficient posterior sample sharing mechanism across particles powered by flow maps. By correcting individual particle drift with the collective posterior samples of the entire ensemble at each resampling step, the framework maximizes sample utility to enable global exploration while actively mitigating reward over-optimization, typical of standard control frameworks. Paired with a principled exploration-exploitation reweighting mechanism involving multi-particle interaction, this sequentially corrected multi-particle dynamics explicitly preserves structural diversity and overcomes the weight degeneracy inherent to standard SMC samplers. Crucially, we prove that the resulting sampling framework yields a multi-particle interaction-aware Feynman-Kac corrector that progressively steers the multi-particle system toward a KL-tilted target distribution, facilitating global exploration and preventing mode collapse. Extensive empirical evaluations and rigorous ablations across diverse search and alignment tasks confirm the efficacy of IMPFM over existing baselines.

0

math.NA 2026-07-02

Monotonic ICNNs approximate polyconvex envelopes with lower cost

by Timo Neumeier, Julian Salmon

Compression of Polyconvex Envelopes of Isotropic Functions via Monotonic Input Convex Neural Networks

A sufficient criterion and positive-octant reduction cut computation while matching the Saint Venant-Kirchhoff envelope closely.

abstract click to expand

This work presents a novel neural-network compression approach for polyconvex envelopes of isotropic functions. The approach relies on a classical sufficient criterion for polyconvexity and is particularly suited for the representation of determinant-constrained energy densities arising in non-linear elasticity. Compared with existing compression methods based on the necessary and sufficient characterisation of polyconvex isotropic functions, the proposed framework reduces computational costs, due to the domain reduction through the restriction to the positive octant in the singed singular value space. The underlying neural-network architecture employs input-convex neural networks (ICNNs) with non-negative weight constraints to enforce the required convexity and monotonicity properties. The additional symmetry and inequality conditions characterising the polyconvex envelope are incorporated weakly through the loss function during training. Although the employed criterion is only sufficient and thus generally yields only a lower bound on the polyconvex envelope, numerical experiments based on the classical Saint Venant--Kirchhoff energy demonstrate that the proposed approach produces accurate approximations in practice while offering a computationally more efficient alternative to existing methods.

0

cs.CE 2026-07-02

Ensemble tops base models for AAA segmentation in CTA

by Joshua Fry, Sajjad Arzemanzadeh +10 more

Stacked Ensemble Learning for Abdominal Aortic Aneurysm Segmentation in CT Angiography

Stacked nnUNet variants reach 0.9752 mean Dice and 0.4598 mm boundary distance on held-out test cases

abstract click to expand

Abdominal aortic aneurysm (AAA) rupture risk assessment increasingly relies on patient-specific biomechanical computations, which require accurate three-dimensional aneurysm geometry from computed tomography angiography (CTA). Manual and semi-automated segmentation remain time-consuming and observer-dependent, limiting their use in large-scale clinical workflows. In this study, we developed a stacked ensemble framework for automated AAA seg-mentation from CTA images. We used 40 anonymised contrast-enhanced CTA scans from AAA patients and generated reference segmentations using the nnInteractive extension in 3D Slicer. We partitioned the dataset into 32 training cases and 8 held-out test cases. Three nnUNetv2 configurations, Default, DA5, and ResEncL, were trained as base learners, and their voxel-wise probability out-puts were combined using an L2-regularised logistic regression meta-model trained from out-of-sample cross-validation predictions. We evaluated segmentation performance using Dice Coefficient and Separation Distance, a mean boundary-to-boundary distance measure introduced in this study to quantify average surface agreement. On the held-out test set, the ensemble achieved the highest mean Dice Coefficient of 0.9752 and the lowest mean Separation Distance of 0.4598 mm, indicating improved volumetric overlap and average boundary agreement compared with the individual base learners. Overall, stacked ensemble learning provided small but meaningful improvements in AAA segmentation, particularly for boundary accuracy relevant to downstream patient-specific bio-mechanical computations.

0

cs.LG 2026-07-02

Entity embeddings tie CatBoost on fraud AUC-ROC

by Xiao Han, Jingjing Liu +3 more

Interpretable vs Learned Encoders for High-Cardinality Fraud Detection

On IEEE-CIS data, embeddings hit 0.9612 while tier encodings stay 0.0064 behind yet keep auditor-friendly boundaries.

abstract click to expand

A total of seven categorical encoding methods were tested on the IEEE-CIS fraud benchmark dataset (590,540 records, 3.5% positives, 8 high-cardinality columns). The encoders were evaluated using a stratified 5-fold cross-validation (CV) with three repetitions. Five of the encoders had identical frozen LightGBM learners in the downstream phase, allowing for controlled comparisons of their performance to each other. CatBoost and TabNet were included as comparisons across paradigms using different learners. The entity embeddings produced the highest AUC-ROC (0.9612), with a statistically significant tie with that of CatBoost (0.9602) and statistically superior to tier group encoding (0.9548), whereas target encoding was only 0.0023 worse than tier group encoding and the auditor-friendly tier boundaries were maintained. Off-the-shelf TabNet did not outperform tree-based pipelines and collapsed under data scarcity. On AUC-PR, CatBoost leads (0.822 vs. 0.793); no encoder dominated both metrics. Per-column analysis confirmed the embedding advantage arises from joint multi-column representation.

0

cs.CE 2026-07-02

Finite-volume bias stabilizes multi-resolution nets for long PDE forecasts

by Xin-Yang Liu, Xiantao Fan +1 more

A Multi-Resolution Finite-Volume Inspired Deep Learning Framework for Spatiotemporal Dynamics Prediction

MuRFiV keeps predictions accurate far into autoregressive rollouts where data-driven networks diverge on Burgers and Navier-Stokes systems.

abstract click to expand

Predicting complex spatiotemporal dynamics in physical processes often demands computationally expensive numerical methods or data-driven neural networks that suffer from high training costs, error accumulation, and limited generalizability to unseen parameters. An effective approach to address these challenges is leveraging physics priors in training neural networks, known as physics-informed deep learning (PiDL). In this work, we introduce the Multi-Resolution Finite-Volume-inspired network, MuRFiV, designed to capitalize on the conservative property of finite volume on the global scale and the expressive power of deep learning on the local scale. We demonstrate the effectiveness of MuRFiV on several spatio-temporal systems governed by partial differential equations (PDEs), including Burgers' equation, shallow water equations, and incompressible Navier-Stokes equations. By embedding PDE information into the deep learning architecture, MuRFiV achieves strong long-term prediction accuracy and remains stable over very long autoregressive rollouts, significantly outperforming data-driven neural network baselines. This result highlights the promise of combining multiresolution learning with finite-volume-inspired inductive bias for accurate and robust long-term prediction of complex dynamics.

0

cs.CE 2026-07-02

10-qubit circuit generalizes 2D training to 3D topology optimization

by Dahyun Joo, Naruethep Sukulthanasorn +2 more

Explainable quantum neural networks for multi-material topology optimization

Mechanical descriptors fed into the quantum network enable material assignment on unseen loads, resolutions, and dimensions without retraini

abstract click to expand

We propose an explainable quantum neural network for multi-material topology optimization, XQNN, that determines both load-carrying structural layout and material type assignment for given boundary/loading conditions. Intermediate solution histories are first converted into element-wise strain energy, sensitivity, density, and Sobel boundary descriptors. Then, they are encoded in a ten-qubit circuit and qubit-wise $Z$ observables are mapped onto material type labels. Trained only on two-dimensional topology optimization histories obtained with a fixed mesh resolution, XQNN can be generalized to handle out-of-distribution boundary/loading conditions, progressively refined high-resolution meshes, and voxel-wise three-dimensional problems without additional training. We find that it is important to preserve qubit-wise observables and add boundary information for improving the optimization accuracy, and certain observables have consistent links to load paths, material type regions, and interfaces, demonstrating their usability as auditable mechanics-facing variables.

0

cs.CE 2026-07-02

Generalized method reaches 100% of Pareto region

by Achille Messac, Blayne Montaque

Generalized Normal Constraint (GNC): A Complete Geometric Generalization of the NNC Method

GNC captures the full admissible set for any number of objectives while NNC and NBI miss a factorial fraction.

abstract click to expand

This paper presents a comprehensive geometric and computational framework for the generation of the complete Pareto frontier. Several existing methods are structurally unable to capture the complete admissible Pareto region. These include widely used methods such as the weighted sum, compromise programming, the Normal Boundary Intersection (NBI) method, and the Normalized Normal Constraint (NNC) method. NNC and NBI, which share the same Pareto-generation grid construction, are structurally unable to capture 50% of the admissible Pareto region for tri-objective problems. More generally, for an n-objective problem, the admissible capture fraction decreases factorially as 1/(n-1)!, and the corresponding missed fraction increases to 1-1/(n-1)!. By contrast, the newly developed Generalized Normal Constraint (GNC) method introduced in this paper is structurally capable of capturing 100% of the admissible Pareto region. The proposed GNC method is formulated for general n-objective optimization problems and is developed through a unified geometric, mathematical, and computational framework supported by insightful examples. Multiobjective optimization plays an important role in a broad range of applications, including economics, product design, and engineering management. Accordingly, the ability of an optimization method to generate a representative subset spanning the complete Pareto frontier is of fundamental importance.

0

cs.LG 2026-07-01

Sparse structure speeds nonlinear projections in physics models

by Alaina Kolli, Theodoros Xenakis +5 more

SNAP-FM: Sparse Nonlinear Accelerated Projection for Physics-Constrained Generative Modeling

Batched local PDE couplings create block-sparse systems that let exact constraint enforcement run faster on linear and nonlinear benchmarks.

abstract click to expand

Generative models have emerged as scalable surrogates for physical simulation, yet they offer no guarantee that their outputs respect the conservation laws, boundary conditions, and nonlinear invariants that govern the underlying physics. Constrained sampling closes this gap, enforcing such constraints exactly at inference time without retraining, but at a computational cost: projection, correction, and trajectory-optimization steps are repeated during sampling, with these steps becoming expensive for nonlinear constraints. Standard ML frameworks exacerbate this: their dense tensor algebra and limited sparse solver composability obscure the structure that physical constraints naturally induce, making efficient batched nonlinear optimization difficult to realize in practice. We address this bottleneck by exploiting the structure that sample-wise batching and local PDE couplings induce in the projection subproblems -- namely, block-sparse Jacobian and KKT systems -- exposing this structure using ExaModels.jl and solving the resulting sparse nonlinear programs with MadNLP.jl and GPU sparse factorization. Applied to Physics-Constrained Flow Matching (PCFM), on PDE benchmarks with linear, nonlinear, one-dimensional, and two-dimensional constraints, this approach accelerates nonlinear constraint projection while maintaining constraint satisfaction. These results show that sparse GPU nonlinear optimization is a practical foundation for constrained generative sampling in scientific machine learning.

0

cs.AI 2026-07-01

Multi-agent LLMs deliver 7.58% CS2 skin returns while market falls 15.62%

by Yao Shi, Kingfung Luo +2 more

CSTrader: A Testbed for Language-Grounded Trading in a Community-Driven Virtual Asset Market

Specialized agents for liquidity, reversed sentiment and friction convert community text into trades that beat both the index and single-pro

abstract click to expand

Niche asset markets, such as Counter-Strike 2 (CS2) weapon skins, are small, volatile, and heavily driven by community discussions and platform rules. These properties make them hard for traditional quantitative models, but provide an ideal testbed for studying how large language models (LLMs) turn unstructured text into trading actions. We present CSTrader, a multi-agent framework for language-grounded trading in the CS2 skin market. The system first integrates heterogeneous signals from various sources, then uses specialized agents for technical analysis, liquidity, events, and (reversed) sentiment, and finally applies risk control, transaction friction, and portfolio management agents to produce buy, sell, or hold decisions under realistic trading frictions. We build a live-like evaluation environment with real CS2 data from a highly volatile period and evaluate several recent LLM backbones. Across models, CSTrader consistently outperforms both a falling market index (-15.62%) and simple single-prompt LLM baselines, achieving up to a 7.58% cumulative return with controlled risk. Ablation studies show that liquidity, reversed sentiment, and transaction friction agents are crucial for turning noisy language signals into stable profits, suggesting that niche, language-driven markets are a useful benchmark for future language-to-action research. Code is available at: https://github.com/IatomicreactorI/CSGOTrading?tab=readme-ov-file#quick-start

0

cs.CE 2026-07-01

Atomic fact hierarchy lifts medical report evaluation

by Yuan Wang, Wanxing Chang +9 more

AtomiMed: Hierarchical Atomic Fact-Checking for Universal Clinical-Aware Medical Report Evaluation

Decomposing reports into diseases and attributes catches diagnostic errors that n-gram metrics overlook and aligns better with radiologists.

abstract click to expand

Traditional metrics for Medical Report Generation (MRG) predominantly rely on surface-level n-gram overlap, which fails to capture clinical factual accuracy and often overlooks catastrophic diagnostic errors. We address this fundamental limitation by proposing \textbf{AtomiMed}, a universal, modality-agnostic evaluation framework that decomposes complex medical narratives into a standardized, multi-level hierarchy of Atomic Clinical Facts, encompassing Disease-level entities and Attribute-level descriptors, including location, morphology, and severity. By implementing an Agentic Cross-Verification loop between ground-truth and predicted reports, AtomiMed simulates a multi-radiologist peer-review process to verify clinical consistency, thus enabling the decoupled assessment of diagnostic detection and descriptive accuracy. To facilitate standardized evaluation, we introduce \textbf{MRGEvalKit}, an open-source toolkit for automated hierarchical extraction, and curate \textbf{OmniMRG-Bench}, a comprehensive multi-modal benchmark covering X-ray, CT, MRI, and Ultrasound. Extensive experiments on multiple expert-annotated reader studies demonstrate that AtomiMed achieves significantly higher correlation with human radiologist judgment compared to traditional and model-based metrics. Our code are release at https://github.com/Venn2336/MRGEvalkit

0

physics.ao-ph 2026-07-01

AI emulator runs global 4.9 km atmosphere after 17 days of training

by Zeyuan Hu, Akshay Subramaniam +10 more

Scaling Storm-Resolving Atmospheric AI Simulation to the Entire Planet

STRATA produces stable 24-hour rollouts at storm-resolving scales with 50 times the energy efficiency of physics models.

abstract click to expand

Kilometer-scale convection shapes precipitation extremes, tropical organization, and cloud feedbacks, but most global atmospheric models approximate these processes at 25-100 km resolution. Global storm-resolving physics models resolve convective systems explicitly, but at a cost -- roughly one MWh per simulated day on exascale supercomputers -- that limits long-duration simulation. We introduce STRATA (Storm-resolving Tile-based autoRegressive Atmosphere Transformer Architecture), the first autoregressive AI emulator for global storm-resolving atmospheric dynamics. STRATA is trained on the highest-resolution atmospheric dataset yet used for global AI emulation: 17 days of SCREAM physics-model output at 4.9-km resolution (~25 million grid cells) sampled every 10 minutes. Our central premise is that on 10-minute timescales atmospheric dynamics are predominantly local, so training on small spatial tiles trades scarce global temporal samples for abundant local spatial samples and enables global rollout via overlapping-tile blending. STRATA combines 3D patch embedding and local 3D neighborhood attention, a novel Stereographic Rotary Position Embedding (StereoRoPE) for grid-invariant encoding, and a pixel-space de-aliasing decoder that suppresses patch-scale rollout artifacts. An iso-FLOP scaling study reveals that km-scale emulation requires ~10x more FLOPs per grid point than coarse-resolution AI weather models, consistent with the higher information density of convective-scale dynamics. Trained on only 17 days of data, STRATA produces stable 24-hour global rollouts with realistic km-scale dynamics across diverse regimes, though large-scale biases develop with lead time. It achieves 48 simulation days per megawatt-hour -- about 50 times better energy efficiency than the SCREAM physics model -- and 741 simulated days per wall-clock day at 512 H100 GPUs. Code and dataset are publicly available.

0

q-bio.BM 2026-06-30

TCR model shows generated structures yield weaker contact maps

by Jiarui Li, Zixiang Yin +5 more

Structure-Regularized Interpretable TCR-Epitope Prediction

TCR-SRIM matches top accuracy yet finds experimental structures produce more accurate and diverse interaction patterns than AlphaFold3 or TC

abstract click to expand

T cell receptor (TCR)-epitope binding prediction is essential for understanding adaptive immunity and developing immunotherapies. Existing sequence- and structure-based models often generalize poorly to unseen epitopes and provide limited interpretability. Furthermore, the impact of generated structures on model learning remains unclear. We present TCR-SRIM, a structure-regularized interpretable-by-design model that combines protein language model embeddings with interpretable contact prototypes to capture residue-level TCR-epitope interactions. TCR-SRIM achieves state-of-the-art predictive performance and improved interpretation quality on the TCR-XAI benchmark. Using its inherent interpretability, we further evaluate the effect of generated structures on model learning. While structures predicted by AlphaFold3, TCRModel2, and tFold-TCR yield competitive performance, they lead to less accurate interaction patterns and reduced binding-site diversity than experimentally-resolved structures. Our results highlight limitations of current structure prediction models for TCR-epitope learning and demonstrate the value of interpretable-by-design models for studying generated biological structures.

0

cs.LG 2026-06-30

Residual matching closes train-test gap in downscaling

by Yujin Kim, Nidhi Soma +1 more

Mind the Residual Gap: Probabilistic Downscaling under Real-World Bias

Optimal transport inside PCA space aligns training residuals to the test regime and cuts under-dispersion on real wind data.

abstract click to expand

Probabilistic downscaling is the task of modeling the conditional distribution of high-resolution fields given coarse inputs, and is a central challenge to atmospheric science, climate modeling, and other multiscale physical systems. A widely used paradigm decomposes the problem into a deterministic mean predictor followed by a stochastic residual generator. While effective in idealized settings, this mean--residual approach frequently produces biased and under-dispersive ensembles in real-world applications. Is this merely generic predictive uncertainty miscalibration? We show that the root cause is more fundamental: residual target misspecification, the residual distribution induced during training differs systematically from the one required at test time due to downscaling bias. To close this gap, we introduce ReMatch (Residual Distribution Matching). ReMatch aligns the training residual distribution toward the test-time regime via optimal transport in a low-dimensional PCA space. This preserves the statistical benefits of the mean--residual framework while reducing the train--test mismatch in the residual targets seen by the stochastic generator. On a controlled synthetic benchmark with varying bias levels and a real-world HRRR--ERA5 wind field downscaling task, ReMatch substantially reduces under-dispersion, improves calibration (SSR and CRPS), and outperforms strong baselines, including the standard mean--residual model and its variants, as well as state-of-the-art super-resolution models. Our code is available at https://github.com/sdean-group/ReMatch.git.

0

cs.CE 2026-06-30

Gradient optimization designs pneumatic soft actuators for target motions

by Anna Dalklint, Vilmer Dahlberg +1 more

Shape optimization of pneumatic soft actuators

Numerical shapes are cast in rubber and their inflation responses match simulation predictions in lab tests.

abstract click to expand

Soft actuators, characterized by their compliance and flexibility, have tremendous potential for diverse applications, ranging from medical devices to submarine operations. However, significant challenges remain in the design of these actuators, specifically in maintaining precise control over their mechanical behavior and motion. To date, heuristic methods have been commonly used to design soft actuators, which are potentially incapable of producing designs that achieve specific target behaviors. We propose a gradient-based inverse design framework to synthesize three dimensional soft actuators with tailored mechanical responses. Our design framework utilizes gradient information that captures the inherent geometrical and material nonlinearities of the soft actuator to morph its shape. We exemplify the capabilities of the proposed framework by designing soft actuators with bespoke deformation patterns, making use of sophisticated deformation mechanisms to realize the target behavior. The capabilities of the proposed framework are validated via experimental testing of cast designs, which confirms a strong correlation between measurements and numerical simulations.

0

cs.CE 2026-06-30

Closed-form basis evolution gives energy-preserving wave ROMs

by Dimitri Goutaudier

Structure-preserving dynamical low-rank approximation for parametric elastic guided waves

For simplified parameters the method removes online integration after loading while exactly conserving energy in dispersive guided-wave simu

abstract click to expand

Elastic guided waves are widely used in Structural Health Monitoring (SHM). In many-query settings, the computational cost of high-fidelity simulations motivates the use of projection-based reduced order modeling (ROM). However, the transport-dominated and dispersive nature of guided waves challenges static linear subspaces. In addition, preserving the Hamiltonian structure of the equations for energy conservation necessitates dedicated projection techniques. While the Dynamical Low Rank Approximation (DLRA) has proven effective for other wave equations, its application to elastic guided waves in SHM has remained unexplored. In this work, we introduce a structure-preserving parametric ROM framework that leverages the DLRA in an off-line/on-line strategy. During the off-line stage, a time-dependent symplectic reduced basis is constructed from training simulations. For a simplified class of parameter dependencies, we derive a closed-form solution of the nonlinear basis evolution equation. This analytical result yields a closed-form, energy-preserving reduced propagator during wave propagation, eliminating on-line time integration after the loading phase. We validate our approach on a 2D elasticity problem featuring dispersive guided waves interacting with a damage. The results demonstrate high compression ratios (rank $\sim 10-30$), low full field reconstruction errors ($\sim 10^{-3}-10^{-2}$), speedups of two to three orders of magnitude, and excellent long-time energy conservation.

0

cs.LG 2026-06-30

Advanced molecular models underperform simpler ones on NMO nano benchmark

by Matthias Blaschke, Daniel Kienzle +4 more

Beyond Drug Discovery: The Nanotechnology Molecular Optimization (NMO) Benchmark

Quantum simulations and strict rules expose limits of drug-pretrained methods while yielding new nanotechnology motifs.

abstract click to expand

Generative molecular design is shaped by simple proxy benchmarks for drug-like properties and models pretrained on large pharmaceutical datasets. This combination yields strong benchmark metrics but limits transferability to domains structurally distinct from drug discovery. To overcome this limitation and drive discovery toward real, scientifically grounded targets, we introduce the Nanotechnology Molecular Optimization (NMO) Benchmark, which bridges machine learning (ML) and quantum materials science. NMO acts simultaneously as a rigorous testbed for the ML community and a discovery engine for nanotechnology research. The suite replaces proxy oracles with quantum simulations and introduces strict protocols that prioritize scientific utility over leaderboard-oriented overfitting. The physics-based NMO tasks impose hard structural constraints and rugged fitness landscapes, posing fundamentally new requirements on generative models. Notably, advanced molecular optimization methods underperform much simpler approaches on the NMO tasks. We develop a new baseline method identifying the critical components to solve the NMO tasks, including a novel representation for modeling structural constraints and a domain-agnostic pretraining strategy to eliminate pharmaceutical dataset bias. Our results surpass state-of-the-art physical properties and reveal previously unknown structural motifs, offering new insights for the nanotechnology community and demonstrating that ML can drive genuine scientific discovery.

0

cs.CR 2026-06-30

Shapley rewards and expander graphs enforce blockchain decentralization

by Yunqi Zhang, Shaileshh Bojja Venkatakrishnan

Rethinking Collaborative Trust for Verifiably Decentralized Blockchain Systems

By tying rewards to collaboration diversity instead of resource uniformity, the design makes decentralization verifiable and tackles scalabi

abstract click to expand

Despite the promise of decentralization, measurement studies have identified a conspicuous lack of decentralization in blockchains. Centralization has been observed in almost all layers of the blockchain, in decentralized applications, and in decentralized autonomous organizations. In many cases, it is practically impossible to definitively determine the extent of centralization in the system. While multiple works have proposed methods to decrease centralization, by and large blockchains continue to be significantly centralized. In this paper, we develop a general framework for building verifiably decentralized blockchain systems. Our framework is motivated by the core observation that the richness and diversity of collaborative interactions between users -- rather than resource uniformity -- captures the essence and extent of decentralization in a blockchain system. Existing blockchains do not have any incentive mechanisms to encourage inter-coalition collaboration, which directly contributes to centralization. We propose a novel reward design that incentivizes users to collaborate with other users without forming isolated coalitions. Technically, our method uses a Sybil-resistant asymmetric Shapley value for reward attribution within a collaboration group, and the theory of expander graphs for measuring and enforcing decentralization. Our framework is general and can be adapted to alleviate centralization in any layer, application, or decentralized organization. It also has important implications beyond the topic of centralization. For example, we show that our solution can naturally address the blockchain scalability problem. We also identify a new class of decentralized collaborative applications that have hitherto been unexplored in blockchains.

0

cs.CL 2026-06-30

Numbers absorbed by next open

by Ding Yu, Zhuo Liu +2 more

Fast Numbers, Slow Language: Bridging Quantitative and Qualitative Earnings Signals

Aligned S&P 1500 data shows quantitative surprises fade fast while qualitative ECT signals generate returns on the following trading day und

abstract click to expand

Earnings announcements release two types of information sequentially: quantitative surprise (numeric earnings-per-share (EPS)/revenue versus analyst estimate) arrives first in press releases and financial news, processed by algorithmic traders within minutes; qualitative language (management tone, guidance, question-and-answer (Q&A) credibility) arrives 30-90 min later in the earnings conference call transcript (ECT), requiring human interpretation overnight. Financial economists have studied quantitative surprise for 50 years; natural language processing (NLP) researchers have studied qualitative ECT signals for a decade. Despite studying the same event, the two communities used incompatible frameworks: different targets (return vs. volatility), trading setups (long top-decile and short bottom-decile vs. trade-all), and metrics (return spread between top and bottom 20% (Q5-Q1) vs. mean squared error (MSE)), making direct comparison and connection challenging. We bridge these communities with EarningsInOne, the first corpus aligning earnings news, ECTs, and intraday and next-day prices across SP 1500 (broad U.S. equity universe, 2022-2025). Applying unified trading and evaluation tools to both signal types, we confirm a clean speed separation, fast numbers, slow language: quantitative surprise peaks at announcement and is largely eliminated by the next market open; qualitative ECT sentiment peaks on the next trading day, real and tradeable, but hidden under prior transcript-based evaluation that optimised sign-agnostic volatility with pointwise MSE.

0

cs.LG 2026-06-29

Physics-guided graph diffusion reaches 0.835% error on device fields

by Yihan Zhang, Zhiteng Zhang +2 more

PCGD: Physics-Guided Conditional Graph Diffusion for TCAD Device Simulation

Hybrid loss on meshes cuts PDE residuals by three orders and adapts to new topologies with 5x less data than full retraining.

abstract click to expand

Technology computer-aided design (TCAD) semiconductor device simulation is fundamentally constrained by the high computational cost of iteratively solving coupled drift-diffusion equations. Existing ML surrogates either reduce internal physics to macroscopic scalar regressions, or rely on single-step mappings that lack the iterative refinement required to resolve stiff, coupled fields. To address this, we introduce PCGD, a Physics-Guided Conditional Graph Diffusion framework operating natively on unstructured TCAD meshes to predict coupled electrostatic and carrier density fields. PCGD employs a Condition-Aware MeshGraphNet denoiser that explicitly injects boundary conditions and device structure context via global cross-attention. By augmenting data-driven denoising with a physics-guided hybrid objective that integrates exponent-free quasi-Fermi gradient matching with noise-aware PDE residuals, PCGD progressively enforce physical constraints in the iterative diffusion trajectory. This strategy successfully bypasses the numerical instabilities typical of stiff drift-diffusion equations. Evaluated on a challenging mixed PN/MOS benchmark, PCGD significantly outperforms deterministic one-step regression (1.207% error) and local diffusion (1.585% error) baselines by achieving a sub-percent mean relative field error of 0.835%, while concurrently reducing maximum PDE residual errors by nearly three orders of magnitude compared to pure diffusion. It also transfers robustly to unseen SOI topologies (0.815% error) via LoRA adaptation, using 5.30$\times$ less data and 14.34$\times$ fewer parameters than full fine-tuning. Ultimately, PCGD bridges the computational efficiency of generative surrogates with the rigorous physical fidelity of traditional TCAD, unlocking highly scalable, field-level analysis for robust device engineering.

0

cs.CE 2026-06-29

AI automates microkinetics discovery with built-in failure recovery

by Ken-ichi Nomura, William Dawson +5 more

Toward Exascale AI for Science: A Scalable AI Skill for Autonomous Microkinetics Discovery

Agentic workflows and surrogate models reduce expert input while checking model reliability for materials research

abstract click to expand

We present a scalable AI-driven framework that advances autonomous scientific discovery by combining agentic workflow automation, high-performance computing, and scientific surrogate models. Using microkinetics discovery as a testbed, the work demonstrates how AI can reduce expert intervention, recover from failed simulations, and systematically evaluate surrogate model reliability. This study shows how AI skills can transform complex domain workflows into robust, scalable capabilities for next-generation materials research.

0

cs.CE 2026-06-29

Integral projection spots fluid regimes in noisy measurements

by Samuel Ahnert, Esther Lagemann +5 more

Weak Dominant Balance for Robust Identification of Dynamically Consistent Fluid Flow Structure

Weak formulation avoids derivatives, succeeds on third-order turbulent equations, and matches simulation with experiment.

abstract click to expand

Extracting interpretable, localized physical mechanisms from complex spatiotemporal data is a foundational challenge across physics, biology, and engineering, but has remained out of reach on real measurements. The central obstacle is obtaining high-quality gradients of data via numerical differentiation, which amplifies noise, diverges for high-order equations, and falters on irregular geometries, limiting the scope of existing approaches to clean simulations of low-order systems. Here, we present weak dominant balance, a derivative-free framework that projects governing equations into a weak (integral) formulation, offloading differentiation onto smooth analytical test functions and leaving the data untouched. The method sustains accurate regime identification under severe noise where existing approaches categorically fail, delivers the first data-driven decomposition of a third-order partial differential equation applied to turbulent duct flow, and produces matching decompositions across direct numerical simulation and particle-image velocimetry measurements of a wavy channel flow, uncovering a previously uncharacterized dynamical regime. Weak dominant balance brings mechanism-level analysis out of simulation and onto measured data, and opens complex physical systems to direct, equation-grounded interpretation.

0

quant-ph 2026-06-29

Quantum method estimates TARF credit risk with 1-8% error

by Sandeep Jha, Richard Oentaryo +8 more

Quantum Counterparty Credit Risk: A Study of Path-Dependent Derivatives

Two-step hybrid uses amplitude estimation after classical conditioning to handle path-dependent FX payoffs on current hardware

abstract click to expand

Estimating potential future exposure (PFE) for path-dependent derivatives, such as FX Target Redemption Forwards (TARFs), represents a formidable computational challenge due to the demand of nested Monte Carlo simulations. We present a hybrid quantum-classical framework that leverages Iterative Quantum Amplitude Estimation (IQAE) to address this via a reduced-order counterparty credit risk model. Our methodology maps the non-linear TARF payoff -- including cumulative gains and knock-out features -- into a quantum circuit via a two-step formulation, whereby a first-step percentile is computed classically and then used to condition quantum evaluation of subsequent exposure. We employ discretisation of the FX process and a linearised additive approximation of dynamics to enable implementation on current quantum platforms. Developed via the Classiq platform and validated on NVIDIA CUDA-Q and Amazon Braket SV1, our approach achieves relative errors of 1%-8% against classical benchmarks at the 97.5% and 99% confidence levels. While discretisation constraints and approximate monotonicity assumption may introduce bias and limit recovery of the full exposure distribution, our framework offers a tractable testbed for quantum acceleration. Scaling analysis suggests that $\sim$300 logical qubits could enable full 52-week exposure estimation, reducing sample complexity for tail-risk estimation via amplitude estimation at the cost of increased circuit depth.

0

math.OC 2026-06-29

Weighted sums miss non-supported points on concave Pareto fronts

by Olaf Frommann

Comparing Scalar Objective Functions for Multi-Criteria Engineering Optimization

Four scalar objective functions tested on analytic convex and concave fronts reveal differences in reachable engineering compromises.

abstract click to expand

Scalar objective functions are required when a multi-criteria optimization problem must yield a single preferred design rather than only a Pareto set. The choice of scalarization influences which compromise is selected, how preference parameters are interpreted, and whether non-supported Pareto regions can be reached. This paper compares four formulations for normalized bi-criteria minimization: weighted sums, achievement scalarizing functions, desirability functions, and a fuzzy-logic-based formulation. Two analytically defined Pareto fronts, one convex and one concave, isolate the effect of the objective formulation from numerical optimizer behavior. The comparison focuses on reachable Pareto regions, parameter-induced selection density, compensation between criteria, sensitivity, and interpretability. Results show that weighted sums are simple but structurally limited on concave fronts, while achievement, desirability, and fuzzy formulations reach interior non-supported regions through different mechanisms. Desirability functions introduce nonlinear single-criterion preference mappings, whereas fuzzy rules express nonseparable and reference-dependent engineering preferences.

0

cs.AI 2026-06-29

AI framework automates novel high-tech system design

by Luuk Oerlemans, Steven Westerhof +1 more

AI-Driven Synthesis for High-Tech System Design: Automating Innovation

Computational design synthesis applies deep learning to move from simulation optimisation to autonomous generation with little human oversig

abstract click to expand

This article addresses the combinatorial complexity inherent in modern high-tech system design by presenting automation-in-design (AiD) as a transformative paradigm. We propose computational design synthesis (CDS), a framework utilising deep learning and generative AI to automate the creation of novel systems. Two case studies (e-drive system design and spatial dimensioning problem) serve as proof-points for this approach. The AI-driven methods used in the case studies represent a fundamental shift in engineering, advancing from simulation-based optimisation towards autonomous design with minimal human supervision.

0

cs.CE 2026-06-29

Single HO-FNO layer beats 16-layer FNO on nonlinear Poisson equation

by Alex Colagrande, Paul Caillon +2 more

Higher-Order Fourier Neural Operator: Explicit Mode Mixer for Nonlinear PDEs

Explicit n-linear Fourier-mode mixing captures polynomial nonlinearities without deep stacking.

abstract click to expand

Neural operators provide deep neural networks for learning mappings between function spaces. Among them, the Fourier Neural Operator (FNO) is particularly effective: its spectral convolution relies on low-dimensional Fourier-domain representations and can handle inputs at different resolutions. This design aligns well with settings where the Fourier basis diagonalizes the underlying operator, such as linear, constant-coefficient PDEs on periodic domains, in which Fourier modes evolve independently. However, nonlinear PDEs may benefit from an additional inductive bias, as they exhibit structured interactions between modes, governed by polynomial nonlinearities. To capture this inductive bias, we introduce the Higher-Order Spectral Convolution, a spectral mixer that extends FNO from diagonal modulation to explicit n-linear mode mixing, aligned with the dynamics of nonlinear PDEs. Our experiments on standard benchmarks show that the proposed Higher-Order FNO (HO-FNO) retains the efficiency of FNO-based architectures and consistently improves over other spectral neural operators. HO-FNO also performs on par with or better than state-of-the-art transformers and state-space models on several datasets, with stronger gains in highly nonlinear regimes, such as the Poisson equation with polynomial forcing, where a single HO-FNO layer outperforms FNO models with up to 16 layers. We open-source our code for reproducibility at: https://github.com/AlexColagrande/HO-FNO.

0

cs.CL 2026-06-29

SMMD loss raises numeric accuracy in LLMs

by Zhuo Zuo, Li Yue +3 more

Enhancing Numerical Prediction in LLMs via Smooth MMD Alignment

Value-distance kernels plus graph smoothness align outputs better than cross-entropy on math, arithmetic, time, and chart tasks.

abstract click to expand

Despite their strong general capabilities, large language models (LLMs) often remain unreliable when outputs must be numerically precise. A key reason is the training objective: standard cross-entropy treats numeric tokens as unstructured categories and ignores the metric structure of their values. We address this mismatch with Smooth Maximum Mean Discrepancy (SMMD), which builds on the classic MMD by incorporating value-distance kernels over numeric tokens and graph-based smoothness. With this kernel defined over a numeric sub-vocabulary, SMMD aligns the predicted numeric distribution to the target via kernel matching and smooths the prediction-target residual over the induced kernel graph to encourage local consistency. We evaluate SMMD on four numeric-target tasks: mathematical reasoning, arithmetic calculation, clock-time recognition, and chart question answering, across multiple open-weight LLM and VLM backbones. SMMD consistently improves accuracy over both cross-entropy and recent numeric-target losses; analyses show complementary effects between MMD and smoothness and underscore the importance of distance-based kernel design. Code is available at https://github.com/Zuozhuo/smmd-loss.

0

math.NA 2026-06-29

Multiplication enforces exact Dirichlet conditions on voxel grids

by Lei Zhang, Jiachen Guo +3 more

Geometry-Preserving Reduced-Order Modeling via Immersed Tensor Decomposition (ITD)

Body-fitted function preserves tensor structure for reduced-order modeling without body-fitted meshes.

abstract click to expand

Body-fitted finite-element methods deliver high-order accuracy but hinge on a clean, watertight, conforming mesh, a requirement that breaks down for the geometrically imperfect CAD assemblies, image-based volumetric data, and voxel-native designs that pervade biomedical engineering and additive manufacturing, where mesh generation has become the dominant cost of the analysis cycle. Immersed methods on regular background Cartesian grids sidestep body-fitted meshing, but classical implementations integrate over irregular cut subdomains, destroying the tensor-product structure that enables separable, reduced-order methods such as tensor decomposition. In this paper we propose the \emph{Immersed Tensor Decomposition} (ITD) framework, which couples a mesh-free geometric representation via body-fitted function with the separable C-HiDeNN-TD reduced-order solver to enable large-scale simulation directly on regular background voxel meshes. The geometry is encoded in three steps: a signed-distance function represents the boundary, a body-fitted function $\Phi$ approximates it with controllable error, and a low-rank Tucker decomposition provides model-order reduction; for a fixed grid spacing $h$, accuracy is improved by raising the approximation order of C-HiDeNN interpolation up to degree $p$ with a linear background mesh. The central contribution is an exact Dirichlet formulation that enforces the boundary condition strongly by multiplying the trial function with $\Phi$, so that $u=g$ holds by construction without any variational penalty or interface quadrature. We establish an a priori error estimate for the formulation and assess it on canonical 2D/3D domains, demonstrating optimal convergence and robustness on non-Cartesian geometries discretized by regular voxel meshes.

0

cs.CE 2026-06-29

Graph attention outperforms time series on crypto prices

by Yu Peng, Matloob Khushi +1 more

CryptoGAT: Are Time Series Models Effective for Cryptocurrency Forecasting?

By modeling cross-asset links as a graph instead of time sequences, it handles volatility that defeats standard temporal models.

abstract click to expand

Cryptocurrency price prediction is a significant challenge in quantitative investment. In recent years, time series models have made significant progress in financial forecasting tasks, especially in the stock market. Despite the growing performance over the past few years, we question the validity of this line of research in cryptocurrency prediction. Specifically, time series models (e.g., LSTM, GRU, and Transformers) are effective at extracting temporal relationships in stock market data. However, in pure price-based cryptocurrency prediction, facing data with extreme volatility and wild swings, time series models have difficulty learning effective information. To validate our claim, we propose CryptoGAT, a lightweight Graph Attention Network that recasts cryptocurrency pure price prediction as a cross-asset graph problem rather than a temporal modeling task. Extensive experiments on real cryptocurrency benchmarks demonstrate that our proposed CryptoGAT outperforms various state-of-the-art forecasting methods with a notable margin. Moreover, we conduct comprehensive empirical studies to explore the fundamental differences exposed by time series models in stock and cryptocurrency prediction: differences in predictability of the signal and cross-asset dependencies. This finding opens up new research directions for the cryptocurrency pure price prediction task and inspires further graph-based exploration in the field. The source code is available at https://github.com/FanBroWell/CryptoGAT

0

cs.CE 2026-06-26

Student team delivers working Doppler tracker for spacecraft

by Ankur Purao, Ella Bianco +3 more

Doppler Tracking of the Artemis II Mission (and Other Spacecraft)

End-to-end S-band receiver and analysis system complete after Artemis II attempt

abstract click to expand

This report describes an American University (AU) student-faculty project to track the Orion spacecraft ``Integrity'' during the Artemis II mission, along its Earth-to-Moon trajectory. The Orion spacecraft has an S-band communications transmitter used for space-to-Earth messages. The S-band transmitter uses an orthogonal quadrature phase shift keying (OQPSK) digital mode, from which Doppler frequency measurements can be derived. This project successfully galvanized student excitement for experimental science, and for space science in particular. Even though we did not successfully obtain a Doppler estimate for the Orion spacecraft, we now have a fully functional end-to-end satellite tracking system as detailed in this report.

0

econ.GN 2026-06-26

REIT concentration links to 2.8 pp higher rent growth

by Advay Ranade

Measuring Racial Disparities in Rent Growth Under Algorithmic Landlord Concentration in U.S. Metros

Association reaches 5.9 points extra in majority-minority tracts across 665 census tracts after AHBI controls.

abstract click to expand

The 2024 Department of Justice antitrust complaint against RealPage, Inc. named five major residential REITs for coordinating algorithmic rent pricing across hundreds of thousands of apartment units in major US metropolitan areas. This paper studies whether census-tract-level corporate landlord concentration (CLC), measured from SEC EDGAR 10-K property filings geocoded to census tracts, the first such application in the literature, is associated with rent growth 2019-2023, and whether that association is larger in majority-minority neighborhoods. Rent outcomes are measured using the Zillow Observed Rent Index (ZORI). To account for the possibility that corporate landlords preferentially locate in neighborhoods already seeing rent appreciation, all regressions control for a fully novel Algorithmic Housing Burden Index (AHBI), a composite of pre-existing rent burden and market tightness from ACS data. Across 665 census tracts in ten US metropolitan areas, doubling REIT concentration is associated with 2.8 percentage points higher rent growth (p = 0.086, p = 0.030, HC1 robust). This association is significantly stronger in majority-minority tracts. Within the same metro, high-CLC majority-minority tracts are associated with 5.9 percentage points higher rent growth than comparable white tracts (p = 0.039). An XGBoost model predicts 44 percent of out-of-sample rent growth variance, with SHAP analysis independently confirming that CLC's contribution is positive in minority tracts and negative in white tracts. Taken all together, these findings provide the first tract-level evidence consistent with corporate landlord concentration being associated with disproportionately higher rent growth in communities of color.

0

cs.CE 2026-06-26

Hybrid solver predicts fracture on new geometries after one training run

by Panos Pantidis, Fouad Amin +2 more

A hybrid IFENN solver for generalizable modeling of phase-field fracture initiation and propagation

Trained once on a benchmark, separate networks for initiation and growth couple with FEM to handle arbitrary shapes.

abstract click to expand

In this paper we demonstrate how the Integrated Finite Element Neural Network (IFENN) framework can effectively model the entire evolution of phase-field fracture, including the initiation and propagation stage, across generalizable geometries. IFENN is a hybrid scheme for coupled computational mechanics problems, tightly coupling a standard FEM solver (mechanical equilibrium) with a pre-trained neural network (coupled field). In this work, the phase-field diffusion equation is approximated with: i) a DeepONet architecture with Kolmogorov-Arnold networks in the trunk and branch (DeepOKAN) for the initiation stage, and ii) a Convolution Neural Network (CNN) for the propagation stage. Both networks are trained only once, on a benchmark geometry, using a purely physics-informed approach based on the maximum strain energy and the phase-field variable. The training process utilizes an extremely small number of training increments and only a limited number of Gauss points that are strategically sampled from the fracture process zone. These features enable a substantial decrease of the offline training cost. To address the extrapolation of the DeepOKAN predictions in regions away from the crack tip during the inference stage, we implement a set of artificial boundary conditions to enforce the near-zero values in the far-field predictions. We showcase the flexibility and numerical accuracy of the proposed methodology across both the training and unseen geometries.

0

cs.CE 2026-06-26

New tensor method accounts for mode interactions to lower error

by Süha Tuna

Holistic Multivariance Decomposition: Adapting Mode Interrelations in Low-Rank Tensor Approximations

HMD uses specialized projections for coupled dimensions, beating Tucker and CP on three benchmarks from different fields.

abstract click to expand

Low-rank tensor approximation is a foundational tool for multidimensional data analysis in scientific computing, classically dominated by Tucker and Canonical Polyadic (CP) decompositions. While widely adopted, these standard approximation schemes represent data as sums of rank-1 tensors formed via mode-wise outer products. This inherent mathematical structure captures the independent variations of individual modes but systematically neglects the mutual interactions and coupled dimensional interdependencies natively embedded within the tensor. To overcome this structural limitation, we introduce the Holistic Multivariance Decomposition (HMD) framework. HMD provides a novel tensor decomposition algorithm that explicitly models both isolated mode effects and higher order mutual relationships through specialized projection operators. Numerical evaluations focusing on three distinct benchmarks from various fields demonstrate that the proposed HMD framework consistently yields significantly lower reconstruction errors compared to both Tucker and CP decomposition. These results establish HMD as a robust, high fidelity computational method for resolving complex, deeply coupled multidimensional data structures in science and engineering applications.

0

cs.CR 2026-06-26

Token standard encodes carrying costs on-chain without rebasing

by JJ Jia Jing Tan, Eva Meng +6 more

The Fungible Reserve Standard: A Deterministic Framework for Encoding Carrying Costs in Asset-Backed Tokens

FRS shrinks asset value per token at a fixed rate then reconciles supply to keep holder balances unchanged.

abstract click to expand

The tokenization of real-world assets (RWAs) has emerged as a transformative application of blockchain technology, with market projections estimating trillions of dollars in tokenized assets within the coming decade. However, a fundamental challenge remains unaddressed: physical assets such as precious metals, stored commodities, and warehoused goods incur structural negative carry -- custody, insurance, and audit costs that accumulate over time. While existing tokenization models have successfully established the market for digital gold and treasuries, they typically manage operational costs at the issuer level. The FRS introduces a framework to bring these economics directly on-chain, avoiding mechanisms such as token rebasing that compromise fungibility and composability with decentralized finance (DeFi) protocols. This paper proposes the Fungible Reserve Standard (FRS), a deterministic token design framework that encodes carrying costs transparently into on-chain logic. The FRS introduces an asset-per-token variable q(t) that decreases according to a predefined annualized carrying cost rate, coupled with a supply reconciliation mechanism that preserves holder balances and ERC-20 composability. While mathematically inspired by the daily expense ratio accrual in traditional asset management -- which often embed centralized profit margins -- the FRS design specifically encodes actual operational carrying costs to provide pure institutional-grade accounting clarity without compromising DeFi compatibility. The framework is asset-agnostic and applicable to any real-world asset with positive, predictable holding costs.

0

cs.CE 2026-06-26

Latent diffusion guides PDE inverse sampling via surrogate

by Yuanzhe Wang, Alexandre M. Tartakovsky

Latent Diffusion Posterior Sampling with Surrogate Likelihood Guidance for PDE Inverse Problems

Method evaluates likelihood gradients through decoder-surrogate composition to avoid repeated full PDE solves on Darcy problem

abstract click to expand

We propose latent-space diffusion posterior sampling (L-DPS), an approximate Bayesian framework for high-dimensional inverse problems governed by partial differential equations (PDEs). The method addresses three challenges in PDE-constrained inversion: implicit sample-based priors without tractable densities, high-dimensional spatially distributed parameters, and the high cost of repeated forward-model evaluations during posterior sampling. L-DPS combines a variational autoencoder, an unconditional latent diffusion model, diffusion posterior sampling, and a differentiable neural surrogate. The VAE maps the parameter field to a lower-dimensional latent space, the diffusion model learns an implicit prior score in this latent space, and DPS combines this learned prior with likelihood-based guidance. The likelihood gradient is evaluated through the decoder-surrogate composition, avoiding repeated calls to the full numerical PDE solver. We evaluate the method on an inverse Darcy flow problem with an unknown spatially distributed permeability field inferred from sparse and noisy pressure observations. L-DPS produces accurate and robust inverse solutions, reduces inference cost relative to full-space DPS, and outperforms amortized inverse baselines such as conditional latent diffusion and inverse FNO in sparse and noisy regimes. We further compare L-DPS with a KLE-MAP baseline and study mixed-prior generalization and the sensitivity of inversion accuracy to surrogate forward-model error.

0

cs.CE 2026-06-26

DPO-tuned LLMs turn ten forecasters into one effective voice

by James Begin, Brendan Gho +7 more

Preference Optimization Drives Monoculture in LLM Prediction Markets

Pairwise error correlations reach 0.70; ten agents match only 1.4 independents and lose to a single model

abstract click to expand

Prediction markets rest on the independence of participant errors. As LLM agents become active traders on platforms like Kalshi and Polymarket, we ask: does this independence hold when the crowd is composed of LLMs? We find it does not. LLM agents fine-tuned with Direct Preference Optimization (DPO) share a convergent output distribution, producing pairwise error correlations of $\rho = 0.70$ and reducing ten agents to the effective forecasting power of ${\approx}1.4$ independent forecasters $N_{\text{eff}}$. This is not a scaling problem: $N_{\text{eff}}$ remains flat from $N=5$ to $N=40$, and the 10-agent market (67.6%) fails to match a single standalone agent (70.2%). Two controlled ablations isolate preference optimization as the causal driver, replicated across labs and scales ($\Delta\rho = +0.24$ to $+0.46$ on identical-SFT controls at 8B and 70B). Among mitigations tested, cross-model diversity achieves the largest correlation reduction ($\rho$ from 0.68 to 0.40). As LLMs become more aligned, markets built from them become more monocultural.

0

cs.LG 2026-06-26

Multi-path model beats global baseline on 8 of 9 bioprocess variables

by Johnny Peng, Thanh Tung Khuat +3 more

Multipath Adaptive Gated Bottleneck Latent ODE with Raman Data Fusion for Cell Culture Process Forecasting

Raman fusion and regime-specific fine-tuning improve forecasts when early bioreactor trajectories diverge across 38 runs.

abstract click to expand

Mammalian cell-culture processes underpin the manufacture of many biopharmaceuticals, yet keeping a run on track is hard: critical process parameters drift over days, and an off-specification trend is often confirmed too late to intervene. Early-stage, multi-day forecasts could enable timely adjustment of feeding, sampling, and control, but bioprocess forecasting is challenging because measurements are sparse and irregularly sampled, operating conditions are heterogeneous across cell lines and media, and runs with near-identical early behaviour can diverge into different futures. We propose an adaptive framework combining a Gated Bottleneck Latent Ordinary Differential Equation (GB-Latent ODE) with Multi-Path Just-In-Time Fine Tuning (MP-JIT-FT). The GB-Latent ODE augments the stan dard Latent ODE with learnable variable-wise gating and a mask-aware bottleneck that compress high-dimensional sparse inputs, improving learning under limited data. Given a partially observed run, MP-JIT-FT retrieves similar historical trajectories, clusters the local neighbourhood into candidate regimes, and fine-tunes a separate model per regime to produce multiple plausible paths, each with a reconstruction-based confidence score, not a single averaged forecast. We further fuse Raman spectroscopy data: a machine-learning soft sensor turns dense Raman spectra into pseudo-observations that enrich the sparse offline measurements for more robust training. On 38 fed-batch 5L bioreactor runs spanning 14 conditions, MP-JIT-FT with Raman fusion achieves the best average rank and outperforms a global Latent ODE baseline on 8 of 9 target variables. Using local-divergence metrics, we show the multi-path gains are largest when locally similar prefixes diverge, whereas Raman fusion helps most when early dynamics are representative of later behaviour.

0

cs.LG 2026-06-25

Otter AI weather model beats NWP baselines with 100x less compute

by Cristiana Diaconu, Jonas Scholz +5 more

Otter Weather: Skillful and Computationally Efficient Medium-Range Weather Forecasting

Deterministic version outperforms traditional forecasts by 9.6 percent at 24 hours after training on under 3.5 A100 days.

abstract click to expand

State-of-the-art medium-range AI weather models can outperform traditional Numerical Weather Prediction (NWP) but require massive training budgets. This restricts usage for under-resourced groups and severely limits fast model iteration. Here we develop Otter Weather, a highly efficient spatiotemporal forecasting model designed to democratise high-performance weather prediction with AI. Evaluated on ERA5 reanalysis data at 1.5{\deg} resolution using standard WeatherBench protocols, the Otter family significantly advances the skill-compute Pareto frontier. The deterministic version outperforms the best NWP baseline by 9.6% at a 24-hour lead time while requiring fewer than 3.5 A100-days for training. It provides a 2x efficiency gain over lightweight AI models and a 100-fold reduction in compute compared to resource-intensive frontier architectures. We extend these efficiency gains into probabilistic forecasting by training via the Continuous Ranked Probability Score (CRPS). Scaling to a larger architecture, Otter-XL achieves a 9.7% CRPS improvement over the IFS ENS baseline. This yields an almost two-fold increase in predictive skill over comparable lightweight models at similar compute budgets. Otter-XL also outperforms frontier architectures like GenCast by over 2%, while using an order of magnitude less compute. Finally, Otter is applied out-of-the-box to a complex acoustic scattering PDE task where it outperforms a state-of-the-art foundation modelling approach, suggesting that the advances made here might apply across a range of scientific domains.

0

cs.DC 2026-06-25

Distributed QAOA matches monolithic results on unit commitment

by Ali Rajabi, Milad Hasanzadeh +1 more

A Distributed Quantum Approximate Optimization Algorithm Simulator for Engineering Design Optimization

The simulator allocates variables across QPUs, handles cross couplings, and recovers identical optimal bitstrings and costs.

abstract click to expand

This paper presents a Qiskit-compatible distributed quantum approximate optimization algorithm (DQAOA) simulator for quadratic unconstrained binary optimization (QUBO) problems arising in engineering design and decision applications. The open-source simulator is available through the RAISE LAB website and GitHub repository, with README documentation for installation, input formatting, configurable parameters, and example workflows. The package addresses the need for a reusable simulator that can solve and compare QUBO instances across different QAOA execution modes. It supports monolithic QAOA on a single quantum processing unit (QPU) and distributed QAOA across a user-specified number of QPUs with configurable capacities. The workflow canonicalizes the QUBO model, maps it to a cost Hamiltonian, allocates variables across QPUs, identifies local and cross-QPU couplings, and constructs the corresponding circuits. Runtime optimizations, including parameterized circuit reuse, objective reuse at fixed depth, batched evaluations, and parallel multi-start execution, reduce repeated overhead. A Streamlit graphical user interface is also provided for entering or uploading QUBO instances, configuring solver settings, running selected modes, and visualizing solution-quality metrics without editing Python scripts. The package is demonstrated on standalone QUBO benchmarks and a power generation unit commitment application. In the unit commitment case, brute force, monolithic QAOA, and distributed QAOA recover the same commitment bitstring and operating cost. Across multiple case studies, the simulator produces results consistent with classical monolithic QAOA references in terms of optimal bitstrings and costs. Staged runtime analysis shows substantial runtime reduction across implementation stages, while distributed QAOA remains more demanding because cross-QPU couplings require remote operations.

0

cs.IR 2026-06-25

Web click embeddings improve financial app recommendations

by Dianjing Fan, Yao Li +5 more

From Clicks to Intent: Cross-Platform Session Embeddings with LLM-Distilled Taxonomy for Financial Services Recommendations

Self-supervised models from pre-login streams raise tile ranking recall and cut log loss while supplying low-latency intent labels.

abstract click to expand

Sequential user behavior modeling is widely adopted in industrial recommender systems; however, significant gaps remain in financial services, where pre-login web interactions and authenticated in-app experiences differ drastically. Specifically, pre-login web users typically explore new products, whereas logged-in app users focus on account servicing. Due to the challenge of cross-channel entity resolution (e.g., matching anonymous web sessions to authenticated mobile accounts), web-based intent signals remain underutilized for post-authentication personalization. Existing methods for capturing web-based intent are often ad-hoc and narrow, lacking the flexibility to support both quantitative downstream recommendations and qualitative understanding at scale. In this work, we propose a scalable and dual-purpose intent prediction framework for web-based interactions and demonstrate its applicability for personalization. Our approach transforms raw web clickstreams into two outputs: a self-supervised Transformer encodes multi-modal clickstreams into a compact session embedding, while an LLM-based taxonomy generation and distillation pipeline produces interpretable intent labels. Our system demonstrates that self-supervised clickstream representations combined with LLM-distilled taxonomies can jointly serve quantitative tasks and qualitative understanding in production: on the mobile homepage tile ranking task, the session embedding improves macro Recall@1 by 1.88% and reduces Log Loss by 13.38% over production baselines. On the user conversion prediction task, the embedding outperforms the LLM labels by 4.3% on micro F1, while the distillation layer delivers interpretable labels at ultra-low latency with only a 7% performance drop.

0

math.OC 2026-06-25

Parametric inventories fold LCA into launcher MDAO

by Alice De Oliveira, Mathieu Balesdent +2 more

A Methodology for Integrating Life Cycle Assessment into a Multidisciplinary Design Analysis and Optimization Framework for Sustainable Launcher Development

Design variables now drive both performance and climate-impact calculations inside a single optimization loop.

abstract click to expand

The increasing number of orbital and sub-orbital launches makes it necessary to investigate the environmental impacts of launch vehicles and incorporate eco-design considerations into their development. In response, the European Space Agency has promoted Life Cycle Assessment (LCA) as a standardization methodology to mitigate environmental impacts of present and future space missions. This need is further amplified in the NewSpace, where numerous configurations and innovative technologies are explored, reinforcing the importance of integrating environmental considerations. At early design stages, launch vehicle architecture can be formalized through a multi-physics optimization problem based on Multidisciplinary Design Analysis and Optimization (MDAO) methods, where disciplines such as propulsion, aerodynamics, structure, and trajectory are coupled to obtain trade-offs among candidate configurations. This paper proposes a methodology to integrate an LCA discipline within an MDAO framework for launch vehicle design. The approach relies on parametric life-cycle inventories depending on design and coupling variables, covering component and propellant production as well as transport to the launch site. Launch emissions are evaluated from optimized trajectory profiles and characterized in terms of climate change impact. The methodology is illustrated on a representative expendable launch vehicle, where multi-objective optimizations assess trade-offs between performance and environmental indicators. Results highlight antagonistic behaviors among environmental impact categories, emphasizing the importance of carefully defining environmental objectives in eco-design studies. The generic nature of the methodology lays the foundation for integrating LCA into early-stage launch vehicle design, enabling exploration of trade-offs between performance, cost, and environmental considerations.

0

q-fin.TR 2026-06-25

Hierarchical graphs boost calendar spread trading in commodity futures

by Yoonsik Hong, Diego Klabjan

Hierarchical Graph Learning for Calendar Spread Strategies in Commodity Futures Markets

Maturity-aware edges raise prediction accuracy and produce positive arbitrage returns on CME contracts.

abstract click to expand

Commodity futures can be represented hierarchically, with underlying assets at the upper level and individual futures contracts at the lower level. Entities at each level can be connected by edges reflecting inherent correlations, with cross-level edges capturing contract-to-underlying asset connections. Building on our observations of these structures, we propose a hierarchical graph learning approach for calendar spread (CS) strategies in commodity futures markets, addressing two significant gaps in the machine-learning literature: (i) the absence of learning-based methods for CS strategies in futures markets, and (ii) the lack of consideration of maturity-dependent interrelationships across commodity futures. We first establish the efficacy of CS strategies by analytically showing that CS strategies can possess higher risk-adjusted returns, measured by the information ratio, and lower risk, measured by variance and delta, than long-only strategies. We then introduce a method to convert learning-based predictions into CS positions. Next, we develop a hierarchical graph learning method that predicts futures price movements by utilizing the maturity-dependent interrelationships, thereby yielding a CS trading algorithm. Empirical results on commodity futures markets traded on the Chicago Mercantile Exchange Group demonstrate that our method outperforms benchmark models in both prediction and trading performance. We find that maturity-dependent interrelationships across commodity futures are instrumental in prediction and that CS trading based on hierarchical graph learning is effective for statistical arbitrage.

0

cs.CE 2026-06-25

ESG long-short portfolios often outperform non-ESG versions

by Giacomo di Tollo, Massimiliano Kaucic +1 more

A Two-Stage Decision Support System for Sustainability-Aware Long Short Portfolio Optimization

Two-stage system adapts asset classes to market conditions then maximizes Omega ratio on 421 European stocks.

abstract click to expand

This paper proposes a two-stage decision support system for long-short portfolio optimization under environmental, social, and governance (ESG) considerations. In the first stage, assets are evaluated using a multi-criteria procedure based on TODIMSort, with criterion weights derived using the MEREC (Removal Effects of Criteria) method. This allows assets to be assigned to classes ordered according to preferences that respond to market conditions and investor priorities, thus generating sets of long and short opportunities that dynamically adapt to the prevailing regime. In the second stage, we formulate a non-convex portfolio optimization problem that maximizes the Omega ratio while respecting budget, bound and leverage constraints. To solve it, we introduce an adaptive particle swarm solver equipped with a controller that selects, at each iteration, the most suitable recombination operator from a diverse pool of operators and combines it with a projection-based repair mechanism for constraint management. The empirical study, conducted on 421 stocks in the STOXX Europe 600 index, examines both the exploration capabilities and solution quality of the proposed solver compared to state-of-the-art benchmarks, as well as the ex post profitability of the resulting portfolio strategies. The results show that ESG-enhanced long-short portfolios offer competitive and often superior performance compared to their non-ESG counterparts and the market-value-weighted benchmark.

0

cs.CE 2026-06-25

Managed grounded LLM delivers advice to island smallholders

by Nikolaos D. Tantaroudas, Ilias Karachalios +1 more

Retrieval-Grounded Multilingual LLM Assistance for Island Smallholder Farmers

A thin proxy with local data retrieval is presented as more attainable and trustworthy than self-hosted models in resource-limited rural set

abstract click to expand

Smallholder farming communities in remote, depopulating areas have limited access to agricultural advice, and their locally specific agronomic knowledge, often expressed in regional dialect, is poorly represented in the global corpora on which Large Language Models (LLMs) are trained. A general-purpose chatbot therefore answers fluently but unreliably, ungrounded in authoritative local data farmers can trust. This paper presents a conversational AI assistant, Falco eleonorae, embedded in a bilingual (Greek-primary, English-secondary) e-market platform serving farmers and cooperatives of a defined island area of interest. It is a thin Backend-for-Frontend (BFF) proxy in front of a geospatially-aware agronomic agent rather than a self-hosted model. Answer generation and tool selection are delegated to a managed upstream service on OpenAI GPT-5-family models, while one bounded task, describing an uploaded field photograph, is handled directly by a vision-capable model so only text reaches the agent, and voice input is transcribed by a managed EU streaming speech-to-text service. Grounding comes not from a self-hosted vector database but from tool-augmented retrieval: a Model Context Protocol (MCP) tool queries a curated, read-only, bilingual data interface exposing local crops, a seasonal calendar, traditional practices, a dialect glossary, products, agritourism experiences, cooperatives, and training content, each wrapped in a geospatial Well-Known Text envelope anchoring the agent to the area of interest. We detail its multilingual, voice, and image modalities, its progressive-web-application and accessibility design for low-bandwidth field use, and its security and data-protection posture, and argue that for a small, resource-constrained rural deployment a managed, grounded multilingual assistant is more attainable and trustworthy than a self-hosted model.

0

cs.CE 2026-06-25

Low-precision cuts HPC resources by 40% in device simulations

by Alexander Maeder, Denghui Lu +7 more

Optimizing Semiconductor Device Simulations through Low-Precision Arithmetic

Stability analysis lets the solver deliver accurate results at 51% higher throughput on less hardware.

abstract click to expand

Architectural changes in GPUs, especially the promotion of low-precision computational units, pose significant challenges to traditional, FP64-based high-performance computing (HPC) applications, while also presenting opportunities. Adopting reduced-precision data formats is a promising avenue to exploit the increased throughput capabilities. However, straightforward data conversions may lead to degraded accuracy or even erroneous results. For a given application, only an in-depth analysis of its numerical stability can reveal the potential of low-precision arithmetic. In this work, we consider the open-source quatrex package, a quantum transport solver capable of breaking the sustained FP64 Eflop/s barrier, to illustrate trade-offs between accuracy losses and computational speed-ups when moving from high- to low-precision formats. We use three representative benchmark structures to explore the application's numerical properties. Applying the gained insights to a larger, more realistic system, we achieve up to 51% higher throughput while maintaining accurate results, on 40% fewer HPC resources than the FP64 reference.

0

physics.flu-dyn 2026-06-25

Neural network speeds vesicle simulations over 100-fold

by Shan Zhong, Gokberk Kabacaoglu +1 more

VesNet: Neural network accelerated solver for simulating Stokesian vesicle suspensions

Approximating local self-interactions while retaining far-field accuracy lets thousands of vesicles run on ordinary computers.

abstract click to expand

Numerical simulation of deformable particle suspensions in Stokes flow is computationally expensive due to nonlinear fluid-structure interactions, evolving interfaces, and multiscale hydrodynamics. We present VesNet, a hybrid framework that accelerates two-dimensional vesicle suspension simulations by approximating vesicle self interactions, including background flow coupling and short-range lubrication forces, while retaining conventional modules for boundary reparameterization and far-field hydrodynamics. A GPU-accelerated implementation achieves over 100x speedup compared to a multithreaded MATLAB CPU boundary integral solver and about 5x relative to its GPU counterpart. VesNet accurately captures key dynamics, including single-vesicle phase behavior, pair interactions, and large-scale suspensions in Taylor-Green and Poiseuille flows, enabling efficient simulations of thousands of vesicles on modest computational resources.

0

math.NA 2026-06-24

3D PDE operators reduce to 1D line kernels without assembly

by Yong Yi Bay, Kathleen A. Yearick

No 3D Matrices: A Unified Tensor-Product View of Matrix-Free Cartesian PDE Solvers

On Cartesian product grids the full three-dimensional matrix is never built, keeping cost linear in total unknowns and storage proportional

abstract click to expand

Every Cartesian three-dimensional PDE solver hides a structural secret that production CFD codes have used for half a century and that graduate-level textbooks rarely state plainly. The derivative matrices, the compact Pad\'e line solves, the Galerkin mass inversions, the alternating-direction-implicit substeps, and even the fast Poisson and Helmholtz diagonalization transforms all factor along the coordinate axes and collapse into repeated one-dimensional banded kernels executed along the grid lines. The three-dimensional operator exists only on paper; it is never assembled, factored, or stored. This paper is the manual for that collapse. We derive the Kronecker-product algebra that makes it exact, carry it cleanly through central differences, compact schemes, tensor-product Galerkin, B-spline and isogeometric methods, collocation, ADI time stepping, and direct Poisson and Helmholtz solves, and bring into the open the three production tricks that turn the reduction into hardware-conscious floating-point throughput on real machines: the multi-right-hand-side reshape that exposes a sweep as one batched line kernel (a dense BLAS-3 GEMM when the line factor is dense or element-local, a banded or stencil kernel when it is not), the sum factorization that rescues high-order Galerkin from the $O(p^{2d})$ quadrature trap, and the pencil decomposition that keeps every direction contiguous across an MPI cluster. For fixed stencil width or fixed polynomial degree, the compute cost stays $O(N)$ in the total number of unknowns $N = N_x N_y N_z$; the operator storage drops to $O(N_x + N_y + N_z)$ up to bandwidth constants; direct separable Poisson and Helmholtz solvers add the expected transform cost; the line kernels are embarrassingly parallel. These facts are familiar to practitioners but rarely assembled in one place; this paper collects them and shows how to use them.

1 0

0

cs.AI 2026-06-24

CoT supervision recovers general knowledge lost during LLM pruning

by Lavinia Ghita, Dhruv Desai +1 more

Scaling Laws for Task-Specific LLM Distillation

Scaling laws from finance domain show supervision format drives tradeoff between task performance and broad capabilities under compression.

abstract click to expand

Large Language Models (LLMs) achieve strong performance across a growing range of domains, yet their scale poses deployment challenges in applications where latency and cost constraints are critical. This paper derives empirical scaling laws for domain-specific LLM compression, quantifying how in-domain and general knowledge performance scale with dataset size, compression ratio, supervision format, and iterative pruning schedule. Using quantitative finance as our application domain, we compare logit-based and LoRA-based distillation under iterative structural pruning, introducing a blended chain-of-thought supervision loss that stabilizes KL-divergence distillation over reasoning traces. In-domain task quality degrades predictably under compression while general-knowledge benchmarks collapse well before the same point; supervision format is the key driver of this tradeoff, with chain-of-thought supervision actively recovering general knowledge that pruning erases. We release the headline dataset FinHeadlineMix, scaling law results, and practical recommendations to provide a reusable framework for domain-specific compression decisions.

0

cs.CE 2026-06-24

Programming model divides composting into two stages

by Zarif Tanzim Aziz, Md. Mahfuzur Rahman +1 more

Development of a Programming Based Kinetic Model for Two Stage Composting of Solid Waste

The approach splits the first 28 days from later degradation to simplify efficiency checks for aeration, moisture, and C/N ratio.

abstract click to expand

As the world is moving toward sustainable development, there is an important need to adopt sustainable waste management solutions of biodegradable solid waste, such as composting, which offers significant advantages over traditional methods like landfilling and incineration by reducing greenhouse gas emissions, enriching soil fertility, and minimizing landfill waste. However, optimizing the composting process governed by factors like aeration, moisture, and carbon-to-nitrogen ratio often relies on complex mathematical models that are difficult to interpret and apply. To overcome this challenge, a user-friendly programming-based two-stage kinetic model has been developed which simplifies composting efficiency analysis, with the first stage covering the initial 28 days of decomposition and the second stage evaluating further degradation, making the process more accessible and actionable for sustainable waste management.

0

cs.CE 2026-06-24

Causal ML plus interpretable models deliver what-if analysis with structure transparency

by David Zapata Gonzalez

A Step Towards Inherently Interpretable Causal Machine Learning Models For Decision Support

Integration for cross-sectional data matches prediction performance while exposing causal links and functional forms.

abstract click to expand

The growing reliance on machine learning for decisions across sectors underscores the importance of model transparency and interpretability. Existing post hoc explainability methods and inherently interpretable approaches shed light on model behavior, yet they primarily reveal how models exploit correlations to maximize performance in prediction tasks. However, many decisions require causal insights and the possibility of using models for what-if scenario evaluation. To address this, we propose the integration of causal machine learning with inherently interpretable models for cross-sectional data. We evaluate these methods in terms of predictive accuracy and interpretability. Our findings show that the proposed approach achieves competitive performance in prediction and what-if analysis while offering transparency on the system structure, causal relationships among variables, and the functional forms that connect them. This work contributes to research on causality, machine learning interpretability, and data-driven decision support by offering informed, transparent, and causally grounded decisions.

0

cs.CE 2026-06-24

Autoencoder addition tests neural reduction on varied vehicle flows

by Kazuto Ando, Rahul Bale +2 more

Neural Network-Based Parametric Model Reduction for Predicting Turbulent Flow for Different Vehicle Geometries

The method reconstructs rear vortices accurately for multiple body shapes in high-speed turbulent air.

abstract click to expand

Numerical simulations in industrial applications often require performing numerous high-precision computations parameterized by specific experimental conditions. For instance, in vehicle body design, aerodynamic simulations are essential for evaluating the aerodynamic characteristics of various proposed body geometries. However, computational resource constraints often become a bottleneck. Therefore, achieving the desired accuracy while minimizing computational cost is crucial. To address this challenge, model reduction methods have been developed to decrease the degrees of freedom by constraining the possible states of a physical system to a lower-dimensional subspace. In particular, reduction techniques that project the system onto a nonlinear subspace using neural networks have been actively studied. Our previous research developed a reduced-order model that integrates neural-network-based model reduction with a time-evolution method, implemented as a distributed parallel training framework to process high-resolution flow field data efficiently. In this study, we extend this reduction approach by incorporating a variational autoencoder to assess its robustness in high-Reynolds-number flows around multiple vehicle bodies with varying geometries. Specifically, we evaluate the reconstruction accuracy of vortex generation across different spatial and temporal scales using a compact latent representation, with a particular focus on the flow behavior near the rear end of the vehicle body.

0

cs.CE 2026-06-23

Graph autoencoders cut cost of sharp-gradient reduced models

by Liam K Magargal, Parisa Khodabakhshi

Efficient implementation of graph autoencoders for model-order reduction of systems with sharp gradients

GNN-LaSDI yields higher accuracy than POD-LaSDI and lower cost than GD-LSPG while adding a metric tuned to gradient locations.

abstract click to expand

This study investigates the efficient deployment of graph autoencoders, a class of graph neural networks (GNNs), for model-order reduction (MOR) of high-dimensional dynamical systems. The proposed framework leverages graph autoencoders to perform nonlinear dimensionality reduction, enabling low-dimensional representations of systems characterized by sharp gradients for which conventional linear approximations, such as proper orthogonal decomposition (POD), are inadequate. Specifically, this study introduces graph neural network latent space dynamics identification (GNN-LaSDI). GNN-LaSDI employs an operator learning framework to directly approximate the temporal evolution of the graph autoencoder's latent representation. The performance of GNN-LaSDI is assessed against both geometric deep least-squares Petrov-Galerkin (GD-LSPG and POD latent space dynamics identification (POD-LaSDI), which combines POD-based dimensionality reduction with operator learning. In addition to standard error metrics, this work presents a novel point cloud error metric specifically tailored to evaluate the accuracy of the identified locations of sharp gradients within the solution. The effectiveness of the metric and the proposed MOR framework is demonstrated through two numerical experiments featuring sharp gradients. For the studied problems, GNN-LaSDI incurs a substantially lower computational cost than GD-LSPG, though it remains slightly more computationally expensive than POD-LaSDI. However, GNN-LaSDI achieves significantly greater accuracy than POD-LaSDI, thereby providing a balance between predictive accuracy and computational speedup. Additionally, the results indicate that the proposed point cloud error provides a more intuitive and informative measure of reduced-order model accuracy in regions with sharp gradients than conventional error metrics.

0

cs.CE 2026-06-23

PyTorch solver recovers G_c from crack patterns with error below 0.001

by Allamaprabhu Ani, Jean-François Molinari +2 more

A matrix-free, differentiable PyTorch solver for phase-field fracture: Formulation, benchmarks, and inverse analysis

Gradient-based L-BFGS optimization inverts the critical energy release rate after only two or three iterations for glass and alumina specime

abstract click to expand

A matrix-free, open-source PyTorch solver is presented for phase-field fracture on central processing units (CPUs) and graphics processing units (GPUs) without custom compiled extensions. In the explicit dynamic pathway, finite-element operations are formulated as element-wise tensor contractions with scatter-based accumulation, removing global sparse mechanics-stiffness assembly from the core time-stepping loop. Both Ambrosio-Tortorelli regularisations (AT1 and AT2), multiple energy decompositions (spectral, volumetric-deviatoric, and star-convex), and plane strain or plane stress assumptions are supported. The explicit mechanics kernels are compatible with PyTorch's automatic differentiation engine (autograd), while the implicit, bound-constrained damage solve is wrapped in a custom backward rule. This rule implements implicit differentiation through the conjugate-gradient (CG) linear solve and keeps memory independent of the internal CG iteration count. The same implementation runs unmodified across macOS, Linux, and Windows, and has been run on meshes of order $10^6$ nodes on a single NVIDIA A100 GPU. The solver is compared against four dynamic fracture cases (straight crack propagation, shear-induced kinking, dynamic branching, and crack-hole interaction in perforated plates) and two quasi-static cases (single-edge notched tension and a notched-holed plate). As a differentiability demonstration, the scalar fracture energy $G_c$ is recovered from observed crack patterns using PyTorch gradients through the forward solve and limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) optimisation. Recovery of $G_c$ with relative error below $10^{-3}$ is achieved after three accepted L-BFGS states for glass and two for alumina. The implementation can be extended and combined with differentiable optimisation and machine-learning components.

0

cs.LG 2026-06-23

LoadKAN matches black-box accuracy while showing mobility effects

by Jinhao Li, Hao Wang

Interpretable Kolmogorov-Arnold Network with Feature-Isolated Temporal Attention Mechanism for Electricity Load Forecasting

Attention isolates features for KAN to expose mobility-load links in three U.S. markets

abstract click to expand

Accurate electricity load forecasting is a crucial prerequisite for stable power system operations. While prevalent deep learning models present competitive performance, they often operate as black boxes and lack interpretability. While the Kolmogorov-Arnold network (KAN) has emerged as a promising alternative because of its learnable activation function design, its direct application to time-series forecasting faces challenges in modeling complex temporal data patterns. Also, simple integration into existing architectures, such as serving as replacement of neural modules, cannot fully leverage KAN's interpretability strengths. To address these gaps, this study develops LoadKAN, a novel hybrid and interpretable framework for load forecasting that synergistically combines a specifically-designed feature-isolated temporal attention mechanism with a KAN module. The attention stage aims to extract temporal dynamics from each input feature independently, such as historical load and human mobility, providing distilled feature representations to the KAN module for interpretable predictions. When evaluated on datasets from three representative U.S. electricity markets, our LoadKAN remains highly competitive when compared to extensively-tuned, state-of-the-art, black-box deep learning benchmarks. More importantly, LoadKAN's interpretability enables a granular analysis of the learned non-linear relationships between six distinct mobility patterns and electricity load. Through KAN-learned activation functions, our quantitative sensitivity analyses on mobility features reveal complex and market-specific dependencies. These findings further demonstrate the ability of our LoadKAN to generate insights often obscured by opaque black-box neural forecasting models.

0

cs.CE 2026-06-23

KDMC hybrid speeds neutral transport 500x with 10% error

by Zhirui Tang, Niels Horsten +1 more

A kinetic-diffusion Monte Carlo-based particle-level fluid-kinetic decomposition for neutral transport simulations

Particle-level fluid-kinetic split avoids coupling iterations and matches AFN accuracy in charge-exchange cases

abstract click to expand

Neutrals in the plasma edge are commonly modeled by kinetic equations, with quantities of interest given by macroscopic quantities such as density, velocity, and temperature. In reactor-relevant regimes, fully kinetic descriptions solved by Monte Carlo (MC) methods, although accurate, become computationally expensive, whereas fluid-limit approximations are computationally more efficient but may lose accuracy due to boundary effects or low-collisional regimes. Hybrid fluid-kinetic approaches aim to combine the strengths of both descriptions. However, existing simulation methods face challenges, including interface handling in domain decomposition, unphysical assumptions, and iterative coupling in distribution decomposition. In this work, we propose a distribution-decomposition hybrid model constructed at the particle level based on the kinetic-diffusion Monte Carlo (KDMC) method. The model inherits key properties of KDMC: it is asymptotic-preserving and does not require iterative coupling between the fluid and kinetic components. To improve the accuracy of the fluid-part quantities estimation, a Navier-Stokes-type fluid system is derived via Hilbert-Chapman-Enskog expansions, tailored for KDMC. In the considered one-dimensional tests, the resulting fluid system has comparable accuracy to the AFN model used in SOLPS-ITER while requiring substantially fewer nonlinear iterations. Additionally, a tunable reflective boundary condition is introduced that allows balancing accuracy and efficiency. The model exhibits at least 500 times speedup over the kinetic MC, while maintaining relative L2 errors around 10% in a charge exchange (CX)-dominant test case. In non-CX-dominant regimes, the accuracy becomes increasingly sensitive to boundary treatment due to the inherent limitations of the fluid approximation near the boundary, motivating further refinement of the KDMC boundary conditions.

0

q-fin.CP 2026-06-23

GPU engine finishes 500 rebalances in 109 seconds

by Debdoot Ghosh

Asymmetry PRISM: A CPU/GPU Portfolio Optimization Engine for Deadline-Bounded Institutional Rebalancing

It meets a 25-minute deadline for a 10,000-instrument universe where the OSQP baseline completes only 4 of 500 accounts.

abstract click to expand

Institutional rebalancing is a batched optimization workload with a hard operating deadline: hundreds of accounts need new weights under budget, turnover, exposure, exclusion, and tax-aware controls before trading can proceed. This paper evaluates Asymmetry PRISM, a CPU/GPU portfolio optimization engine, through a public evaluation boundary; problem data in, and returned weights, status codes, timings, memory class, external feasibility diagnostics, eligible objective comparisons, and audit records out. Within that boundary, the evaluation protocol fixes hardware and software versions, declares timing lanes, separates cold single calls from repeated workloads, and admits objective-gap claims only where an eligible reference solver completed. On completed multi-solver rows from N=100 to N=2,000, Asymmetry PRISM-CPU is 4.5x to 24.1x faster than the fastest completed reference row in the same lane. In the production queue study, Asymmetry PRISM-GPU completes 500/500 accounts over a 10,000-instrument universe in 109.5 s within a declared 25-minute operating window, with zero missed deadlines and an audit record for every solve; the recorded OSQP queue baseline completes 4/500. On an operationally constrained real-data suite (tax-motivated transition penalties, restriction caps, turnover controls, batches), Asymmetry PRISM clears constrained solves 3.4x to 126.7x faster than the best completing incumbent at certified-equal objectives, and the GPU route widens to 8.8x over the CPU route at N=384,800. Rows without a completed reference are reported as feasibility, timing, memory, and failure-status evidence.

0

cs.CE 2026-06-23

Self-attention surrogate predicts free-surface flows scalably

by Federico Lanteri, Massimiliano Cremonesi

Attention mechanism for scalable mesh-based neural surrogates of free-surface fluids

It keeps the PFEM mesh structure intact for accurate long-term dynamics and stress calculations in engineering flows.

abstract click to expand

High-fidelity simulations of free-surface flows using Lagrangian methods such as the Particle Finite Element Method (PFEM) are computationally demanding due to continuous domain updates and repeated solution of the governing equations. This challenge is further amplified by non-Newtonian rheologies, where material nonlinearities increase computational cost. These limitations motivate the development of efficient surrogate models to approximate PFEM dynamics at reduced cost. While data-driven deep learning approaches are promising, a key challenge is designing models that operate on arbitrary and evolving geometries. We propose a self-attention-based neural surrogate for PFEM simulations of free-surface flows. The architecture leverages attention mechanisms to model node interactions and capture complex spatial dependencies, while preserving the PFEM mesh discretization. This provides a geometric and topological framework for remeshing and node redistribution, maintaining high-quality spatial discretization during rollouts, improving long-term stability, and enabling reconstruction of derived mechanical quantities via standard finite element operators. Two attention formulations are considered: a standard self-attention mechanism and a linear variant that reduces computational cost and improves scalability. The models are evaluated on two- and three-dimensional free-surface flow benchmarks with evolving geometries, varying material parameters, and non-Newtonian fluids. Results show accurate prediction of transient dynamics and final configurations, with significantly improved scalability. The mesh-based formulation also enables direct reconstruction of quantities such as stress fields. Overall, the framework provides an accurate and scalable surrogate strategy for PFEM simulations in engineering-scale applications.

0

cs.CE 2026-06-23

Two-layer method values data assets under IAS 38

by Natasha E. Blycha, James Myint +8 more

Pricing the Unpriced Asset: A Standards-Based Method for Valuing Enterprise Data under IAS 38 and IAS 2

Cost-based D-Val and commercial A-Val give authenticated datasets measurable worth before active markets form.

abstract click to expand

The recognition and measurement of data assets under current accounting standards presents significant challenges. While International Accounting Standard 38 (IAS 38) provides a framework for intangible asset recognition, data assets frequently fail to meet capitalisation criteria due to difficulties in demonstrating separability, establishing reliable cost measurement, and proving probable future economic benefits. The widespread failure to easily and reliably value data causes mispricing and allocative distortions across data and artificial intelligence markets. This paper introduces a two-layer valuation progression for authenticated data assets, that is, datasets that have met IAS 38 recognition criteria through established legal provenance and contractual boundaries. The first layer, D-Val, is the auditable cost-basis valuation consistent with IAS 38. D-Val is defined as D-Val = Cp * Avt, where Cp is the reliably measurable production cost and Avt is the appreciation or depreciation factor applied over time. Under prevailing interpretations of IAS 38, Av is constrained to values less than or equal to one absent an active market revaluation, rendering D-Val a strictly cost-less-amortisation figure. The second layer, A-Val, is a theoretically grounded commercial valuation that incorporates scarcity, rivalry, completeness, accuracy, and explicit premia for provenance authentication and independent audit. A-Val is not auditable as fair value under current practice but serves as a defensible commercial valuation during the period before active markets for authenticated data assets mature. As authenticated data markets mature parameter assumptions improve providing a foundation for iterative refinement of the model.

0

cs.LG 2026-06-23

Flow matching produces path-dependent stress fields in one step

by Yijing Zhou, Jasmin Jelovica

One-Step Flow Matching for Generative Modeling of Path-Dependent Physical Fields

Transformer model in VAE space reaches FEM accuracy with 6-7x CPU and ~100x GPU speedup using modest training data

abstract click to expand

Physical simulations for intricate geometries with path-dependent constitutive models face difficulties due to the enormous computational cost they require. Recently, the emergence of generative AI models, which succeed in image and video synthesis tasks, has provided a promise to further improve simulations. Although U-Net-based denoising diffusion probabilistic models (DDPMs) have been adopted for elastic stress field generation, they typically require hundreds of sampling steps, and applications of generative models to path-dependent, e.g. plastic, stress fields remain very limited. In this work, we propose a novel flow matching (FM) model based on a transformer backbone for high-resolution path-dependent stress field generation with stochastic loading-unloading paths and geometry. The proposed model operates within the latent space of a variational autoencoder (VAE) and formulates the simulation of plastic fields as a video synthesis task, directly generating the stress fields across all time steps. Meanwhile, we design a non-Gaussian source distribution for flow matching, such that crossings among conditional transport paths are reduced during training. This enables our model to generate satisfactory samples in one step without relying on distillation. In addition, we introduce token-level loading embeddings and two auxiliary networks to further enhance the model performance in path-dependent simulation. The results demonstrate that, even with a limited training dataset, our model can accurately generate high-resolution path-dependent fields. It is much more computationally efficient than finite element analysis, providing a speedup of 6 to 7 times over FEM on CPUs and approximately two orders of magnitude speedup on consumer-grade GPUs.

0

cs.CE 2026-06-22

Hybrid model predicts monetary relief in complaints at 0.78 AUC

by Zhuoer Wang, Sizhen Zhu +1 more

From Complaint Narratives to Monetary Relief: A Hybrid Machine Learning Framework for CFPB Consumer Complaints

Combining narratives, LDA topics and company data lifts performance over TF-IDF baseline on temporal CFPB split.

abstract click to expand

Consumer financial complaints provide a valuable source of information for identifying service failures, dispute frictions, and operational deficiencies in consumer-facing financial institutions. This paper proposes a hybrid machine learning framework for predicting monetary relief outcomes using Consumer Financial Protection Bureau complaint data. We formulate the task as an imbalanced binary classification problem, where complaints closed with monetary relief are treated as compensable outcomes. The proposed framework integrates multiple sources of predictive information, including complaint narrative text, LDA-based topic representations, interpretable text-engineered features, and structured categorical attributes such as company and state. An XGBoost classifier is trained using a temporal train-test split, with earlier complaints used for model development and more recent complaints reserved for out-of-sample evaluation. Compared with a TF-IDF baseline, the proposed framework substantially improves predictive performance, increasing AUC-ROC from 0.69 to 0.78 and improving PR-AUC under class imbalance. Feature importance analysis shows that textual signals, latent complaint topics, and company identity all contribute meaningful predictive information. In particular, company-level effects reveal systematic variation in complaint resolution patterns across financial institutions. These findings suggest that consumer complaint narratives can serve as alternative data for monitoring consumer harm, identifying firm-level operational weaknesses, and supporting early-stage risk surveillance in consumer finance.

0

cs.CE 2026-06-22

Phase-field model predicts MIC pitting rates under SRB and stress

by S. Kovacevic, E. Martínez-Pañeda

A phase-field model for microbiologically influenced corrosion

The formulation links Monod microbial kinetics and stress-enhanced mobility to forecast damage from microstructure to monopile scales, inclu

abstract click to expand

A phase-field-based reaction-diffusion corrosion model is developed to predict microbially influenced corrosion (MIC) in metal alloys, with a focus on anaerobic conditions and sulfate-reducing bacteria (SRB). The formulation couples microbial sulfate reduction, sulfate transport, electrochemical kinetics, material dissolution, and mechanical effects. Microbial activity is modelled using a Monod-type expression for sulfate consumption, whereas the mechano-chemical coupling is incorporated through an enhanced mobility relationship that captures the influence of mechanical fields on corrosion kinetics. The model is calibrated against experiments and shows strong agreement in predicting pitting kinetics under SRB activity. Sensitivity analyses quantify the competing roles of microbial kinetics, transport, and thermodynamic driving forces in governing corrosion behaviour. The capability of the formulation to capture both MIC-induced pitting and stress-assisted corrosion across multiple length scales is demonstrated through case studies that include microstructure-sensitive simulations and structural-scale coupling with a cathodic protection (CP) model. Results show that finer grain sizes reduce pitting severity but promote faster defect propagation under mechanical loading. At the structural scale, coupling with the CP model enables predictions of defect growth under varying electrochemical conditions and over engineering-relevant length scales, as exemplified with the analysis of an offshore wind turbine monopile. CP delays pitting and suppresses crack propagation, although its effectiveness diminishes as sacrificial anodes degrade. The framework provides a predictive and computationally efficient tool for assessing MIC-induced damage over extended times, with potential applications in the integrity and life assessment of metallic structures operating in aggressive microbial environments.

0

cs.CE 2026-06-22

Network surrogate cuts piezoelectric homogenization cost 1000-fold

by Ting-Ju Wei, Yen-Ming Lu +1 more

Deep material network for homogenization of piezoelectric composites

Embedding coupling laws lets a model trained on linear data predict nonlinear responses with DNS-level accuracy.

abstract click to expand

Piezoelectric composites are widely used in sensors, actuators, transducers, and energy-harvesting devices because their effective electromechanical performance can be tailored by combining constituent phases and microstructural architecture. However, conventional computational homogenization based on direct numerical simulation (DNS) is computationally expensive, particularly for multiscale simulations and material design tasks that require repeated homogenization analyses. To address this limitation, this work proposes a piezoelectric deep material network (PDMN) to efficiently homogenize two-phase piezoelectric composites. The proposed framework embeds the governing electromechanical homogenization relations directly into the network architecture, yielding a physics-informed, semi-analytical surrogate that explicitly captures the two-way coupling between the mechanical and electrical fields across constituent phases. The network is trained offline on linear electroelastic datasets and, through a fully coupled Newton--Raphson solution with a consistent electromechanical tangent, subsequently used for efficient online prediction under broader constitutive settings, including nonlinear electroelasticity and history-dependent responses. The framework is validated on two-phase composites of polyvinylidene fluoride (PVDF) and lithium niobate (LiNbO$_3$) with reversed phase arrangements under nonlinear electroelastic loading, and on a viscoelastic--piezoelectric composite exhibiting coupled stress relaxation. Numerical examples show that the proposed PDMN achieves high predictive accuracy while reducing the computational cost by more than three orders of magnitude compared with DNS. The proposed framework, therefore, provides an efficient and reliable surrogate for the multiscale analysis and design of piezoelectric composites.

0

cs.CE 2026-06-22

Hybrid method designs mmWave inductors in minutes

by Yuzhen Song, Yifan Wang +6 more

HFORD: Hybrid Forward Optimization and Reverse Design Method and Its Applications to On-Chip Millimeter-Wave Inductive Elements

HFORD combines ML models and optimization to map targets to DRC-compliant layout seeds for on-chip devices.

abstract click to expand

On-chip inductive elements are pivotal in determining both the silicon footprint and performance of millimeter-wave (mmWave) integrated circuits. However, the layout-level synthesis of these passive devices is severely challenged by highly nonlinear geometry-to-performance mappings, computationally expensive full-wave electromagnetic simulations, topology-dependent design spaces, and the inherent non-uniqueness of inverse design. To overcome these bottlenecks, we propose a hybrid forward optimization and reverse design (HFORD) method for the target-to-layout synthesis of mmWave inductive elements. Utilizing a unified core to map device-level requirements to layout-level seeds, HFORD structures direct device targets and translates circuit specifications into a hierarchical synthesis flow. Specifically, sparse-fitting sampling is introduced to improve coverage across critical performance regions, while compact response-fitting coefficients significantly reduce training dimensionality. The HFORD core integrates a random forest for topology selection, a variational autoencoder for spectral feature generation, a mixture density network for probabilistic inverse mapping, and particle swarm optimization for latent space exploration. This integration improves the feasibility of the generated layout seeds under design rule check (DRC) constraints. Two design examples demonstrate that the proposed method accelerates the design cycle from hours to minutes compared to conventional optimization methods.

0

cs.AI 2026-06-22

Simulation rollouts teach compact models to pick winning market strategies

by Jiaxiang Chen, Aotian Luo +4 more

MetaPS: Adaptive Programmatic Strategy Selection for Market Agents

MetaPS outperforms direct LLM agents and fixed strategies by learning state-dependent program selection from backtests.

abstract click to expand

No single market strategy always wins: momentum, mean reversion, risk control,and event-driven rules can each succeed or fail as market conditions change.Rather than asking large language models to directly generate market actions,we study an executable decision paradigm where an agent selects from a library of programmatic strategies, each implemented as a code module mapping market observations to actions.We propose \textbf{MetaPS}, a simulation-guided framework for adaptive programmatic strategy selection. MetaPS rolls out candidate strategies in simulated or backtested markets, identifies states where particular strategies lead to better future outcomes, and converts these state--strategy pairs into supervised fine-tuning data. During inference, the simulator is no longer queried: MetaPS observes only the current market state and candidate strategy context, selects a suitable strategy program, and the selected program produces the final action. Experiments on multi-stock trading and a controlled goods-exchange sandbox show that MetaPS consistently improves across model scales from 0.8B to 9B parameters. It outperforms fixed-strategy baselines, direct decision-making agents, and prompted API-based LLM agents; in several settings, compact fine-tuned models even surpass stronger API models. These results demonstrate that market simulations can provide scalable and targeted supervision for learning adaptive, interpretable, and executable strategy selection.

0

cs.AI 2026-06-22

Three causal tiers fix LLM over-anchoring in materials discovery

by Yi Cao, Liaoyaqi Wang +4 more

ARIA: A Causal-Aware Framework for Rescuing LLM Reasoning in Trustworthy Materials Discovery

ARIA routes queries by completeness of process-structure-property chains, adding traceable physics reasoning and lifting baseline performanc

abstract click to expand

Generative models have revolutionized the process of materials discovery, yet they often fail to satisfy underlying physical causality. Through an analysis of Large Language Models (LLMs) augmented with knowledge graphs derived from current literature, we uncover a phenomenon termed contextual tunneling, where models "over-anchor" on narrow, retrieved evidence while suppressing global physical reasoning. To address this problem, we introduce ARIA, a causal-aware framework that conditions knowledge use on mechanistic completeness. ARIA routes each query through a three-tier cascade: (i) direct causal reasoning when complete evidence chains of Process-Structure-Property (PSP) are available, (ii) physics-informed analogical transfer for sparse or novel material systems, and (iii) explicit parametric fallback when external evidence is incomplete. As a proof of concept, we construct a Knowledge Graph (KG) containing 2,839 extracted PSP relations from peer-reviewed articles in the materials literature and evaluate ARIA on forward prediction and inverse design tasks for two-dimensional (2D) materials. ARIA mitigates contextual tunneling, improves over unaugmented and naive KG-augmented baselines, and provides further gains when an online literature search is used for evidence enrichment. Crucially, ARIA produces auditable causal traces, enabling physically grounded and trustworthy AI-assisted materials discovery.

0

cs.CE 2026-06-22

Fare-free buses raise low-income ridership but cut revenue in NYC model

by Parker Wischhover, Hossein Amiri +5 more

Simulating Public Transit Fare Policies in NYC: An Efficient, Socioeconomic-Aware Framework

Simulation shows small total ridership change yet clear shifts in bus versus subway use and income-group effects

abstract click to expand

Designing equitable and effective public transit fare policies is challenging due to complex interactions among traveler behavior, multimodal networks, and socioeconomic heterogeneity. This paper presents a scalable, data-driven simulation framework for evaluating transit fare policies in New York City (NYC), integrating a synthetic population, agent-based simulation, multimodal travel-time estimation, and fare-sensitive mode choice modeling. We evaluate multiple fare scenarios, including distance-based pricing, fare increases, and fare-free bus policies. Results show that pricing changes modestly affect total ridership but significantly alter modal composition and produce heterogeneous impacts across income groups. In particular, fare-free bus policies generate substantial benefits for lower-income riders by increasing bus usage and reducing fare burden, while introducing trade-offs in revenue. To support city-scale analysis, we introduce a sampling-based approach that reduces computational cost while preserving aggregate accuracy. The proposed framework provides a practical tool for assessing trade-offs between ridership, revenue, and equity, enabling more informed and equitable transit policy design.

0

cs.CE 2026-06-22

Differentiable model accelerates adsorption design 100-fold

by Alex Glover, Maria M. Papathanasiou +1 more

Accelerating Simulation and Optimisation of Cyclic Adsorption Processes with Differentiable Programming

Exact cycle Jacobians cut cyclic steady state time by 20x and Pareto optimization by two orders of magnitude while keeping the full mechanis

abstract click to expand

The design of cyclic adsorption processes is computationally demanding, requiring repeated convergence to cyclic steady state within an iterative optimisation loop. Conventional workflows treat the process simulator as a black box and rely on derivative-free optimisation, resulting in design campaigns that can require hundreds to thousands of CPU hours. This work presents an end-to-end differentiable model of a pressure vacuum swing adsorption process, developed using the JAX differentiable programming framework and applied here to a benchmark post-combustion carbon capture problem. Automatic differentiation provides exact gradients throughout the entire computational workflow. The differentiation of a single process cycle provides the Jacobian for a Newton iteration to decrease both the number of iterations and the simulation time required to reach cyclic steady state by a factor of 20 relative to a representative MATLAB implementation. Exact gradients of the performance metrics with respect to the design variables further enable gradient-based multi-objective optimisation using the IPOPT algorithm. Applied to a six-variable design problem, the latter produces a superior Pareto front with improved coverage of the trade-off space and closer convergence to the optimal front than the genetic algorithm NSGA-II. Notably, the full front is obtained over two orders of magnitude faster than the conventional approach. By retaining the full mechanistic model while making it differentiable, this framework transforms cyclic adsorption process design from slow black-box simulation with derivative-free optimisation to efficient gradient-enhanced modelling and optimisation, enabling rapid and systematic exploration of complex design spaces.

0

cs.CE 2026-06-22

Tool aligns 11 wildfire datasets into ready NumPy arrays

by Zeyu Xia, Lexie Chen +2 more

FireDataForge: A Unified Framework for Multi-Source Wildfire Data Retrieval and Integration

FireDataForge takes one event ID and outputs harmonized data from fire behavior to land cover for analysis.

abstract click to expand

Wildfire research, modeling, and education require geospatial data from multiple sources that vary in formats, coordinate systems, spatial resolutions, and temporal cadences. This preprocessing burden limits reproducible reuse. We present FireDataForge, an open-source Python framework that automates retrieval and harmonization of 11 wildfire-related sources spanning fire behavior, weather, land cover, vegetation, elevation, built environment, wildland-urban interface, fire history, and satellite imagery. Given an MTBS Event ID, FireDataForge retrieves relevant datasets, aligns them to a common grid, and outputs analysis-ready NumPy arrays with embedded metadata. Batch processing of historical fires demonstrates support for fire behavior simulation, educational visualization, machine learning, and AI-assisted wildfire analysis.

0

cs.CE 2026-06-22

This pilot study benchmarked five EEG functional connectivity metrics for decoding action…

by Anh T. Nguyen, Zachary A. Rentala +1 more

Robust EEG Functional Connectivity Metrics for Decoding Action Observation Conditions and Observed Actions

Pilot data show imaginary coherence and directed measures outperform standard coherence when classifying observed actions with GNNs.

abstract click to expand

Action observation (AO) paradigms probe motor-system engagement, yet the electroencephalographic (EEG) functional connectivity (FC) metrics that best capture AO-related dynamics remain unclear. This pilot study benchmarked five sensor-level FC metrics, including coherence (COH), imaginary coherence (iCOH), phase-locking value (PLV), partial directed coherence (PDC), and spectral Granger causality (SpcG), for decoding AO stimuli in five healthy adults. EEG signals were recorded while participants observed upper-limb actions performed by human or robot agents, as well as non-action control stimuli. Ten motor-area channels were analyzed in the alpha (8-12 Hz) and beta (13-30 Hz) bands. Trial-wise 10 x 10 FC matrices were used as inputs to multiple classifiers for two tasks: (i) six-class AO-condition decoding and (ii) five-class action-type decoding. Across both tasks, metrics robust to volume conduction consistently outperformed their counterparts. iCOH achieved the highest macro-area under the receiver operating characteristic curve (macro-AUC) for most classifiers, with PDC and SpcG showing comparable performance. Graph neural networks (GNNs) provided the most robust and stable results across all FC metrics, while convolutional neural networks and random forests also performed strongly. These findings highlight the importance of suppressing zero-phase-lag interactions and incorporating directed connectivity when characterizing AO-related brain activity. They further demonstrate the ability of GNNs to exploit the inherent graph structure of FC representations, providing practical guidance for selecting connectivity measures and machine learning models in future large-scale studies of action observation.

0

cs.LG 2026-06-19

Dual-stream graph model predicts panel stress fields with fewer samples

by Yuecheng Cai, Jasmin Jelovica

Physics-Guided Dual-Stream Heterogeneous Graph Neural Network for Predicting Full-Field Structural Response of Stiffened Panels

DS-HGNN records lowest RMSE versus six benchmarks and matches top accuracy using 19-38 percent less training data while capturing yield beha

abstract click to expand

Iterative design and optimization of large, complex structures require fast and accurate prediction of stress, displacement, and other fields. Finite element analysis (FEA) is computationally expensive for this task. Existing neural network surrogates often struggle with varying topologies and complex boundary conditions. This study proposes the novel Dual-Stream Heterogeneous Graph Neural Network (DS-HGNN) for full-field stress and displacement prediction in thin-walled structures, demonstrated on box beams made of stiffened panels. DS-HGNN operates on panel-level heterogeneous graph representations and introduces physics-guided edge states initialized from edge types, spatial information, and boundary kinematics. These states are updated through dual-stream message passing that separates longitudinal and transverse structural information while allowing cross-stream exchange. Geometry and loading effects are incorporated through Feature-wise Linear Modulation (FiLM)-conditioned 1-D spectral convolutions, and physical fields are reconstructed using a spectral-bypass low-rank readout. The model is evaluated on stiffened panel datasets with different geometries, boundary kinematics, loading conditions, and material nonlinear responses. DS-HGNN achieves the lowest stress and displacement RMSE compared with six benchmark heterogeneous graph neural network models. It also reaches comparable accuracy to the strongest benchmark models using 19%-38% fewer training samples. A targeted evaluation further shows that DS-HGNN captures yield and post-yield stress features.

0

cs.CE 2026-06-19

Meta-learning cuts Pareto error 78% in spin-crossover search

by Antonio Varagnolo, Yulia Pimonova +3 more

Interpretable Meta-Learning for Multi-Objective Chemical Search

Linear models trained on auxiliary properties adapt to new competing objectives from limited data.

abstract click to expand

Navigating the vast space of synthetically accessible molecules demands surrogate models that are interpretable and capable of handling multiple competing objectives at the same time. Deep learning approaches struggle to satisfy them under the computational constraints of quantum-level chemistry. Here, we introduce a modular pipeline that combines interpretable linear meta-learning models and adaptive-confidence uncertainty quantification into an Efficient Global Optimization (EGO) framework for multi-objective molecular discovery. For the first time, linear meta-learning is deployed in a multi-objective chemical search setting: by training across chemical objectives and cheap auxiliary properties, the meta-learned surrogates acquire transferable chemical knowledge that adapts rapidly to new objectives from limited data. Evaluated empirically on a real large scale search for spin-crossover metal-organic complexes, the baseline performs 78% worse in Pareto sense than the meta-learning alternative. We also address the calibration challenges inherent to active search. Since optimal candidates typically lie precisely in the distributional tails, standard uncertainty estimates fail. We introduce an adaptive confidence-tuning algorithm that dynamically recalibrates the exploration-exploitation trade-off as the molecular search evolves. Empirically, dynamic confidence tuning further dominates over 50% of the statically calibrated front.

0

cond-mat.mtrl-sci 2026-06-19

Tessellation BCs create single clear localization bands in concrete models

by Jan Raisinger, Jan Eliáš

On representation of macroscopic crack in periodic fine-scale discrete mechanical models

Percolation-path-aligned conditions produce multiple or spurious bands; tessellation ones set band length solely by geometry.

abstract click to expand

In multiscale modeling of heterogeneous softening materials, boundary conditions (BC) in the fine-scale model strongly influence the strain localization pattern and the macroscopic response. For rectilinear models (e.g., squares or cubes), standard Periodic BCs produce artificially ductile behavior with excessive energy dissipation when the localization band inclination does not match the periodicity directions. Recently proposed Tessellation and Percolation-path-aligned BCs promise to address this by adapting the periodicity frame to align with the evolving localization bands. Alternatively, spherical/circular models provide an orientation independent response by design. Unfortunately, the standard Periodic BCs do not allow development of proper localization band crossing spherical model's boundaries. A recently proposed modification addresses this by adding a displacement jump to the spherical periodic BCs. This study evaluates the applicability of these novel BCs to a mesoscale discrete particle model of concrete. Two-dimensional square and circular models under uniaxial tension with different loading directions are analyzed, with the selected approaches extended to three-dimensional cube models. Results show that Percolation-path-aligned BCs exhibit major shortcomings: they can lead to multiple localization bands due to uneven straining of the two boundary sections and their weakly constrained section can be prone to spurious strain localization. In contrast, Tessellation BCs consistently yield a well-defined localization band, whose length is determined solely by the model geometry, making it straightforward to account for in post-processing. Periodic boundary conditions augmented with a displacement jump applied to a circular model sometimes incorrect produce crack patterns similar to those under the standard Periodic BCs.

0

quant-ph 2026-06-19

Learned circuits beat Trotter scaling for quantum operators

by Paul Over, Sergio Bengoechea +4 more

Operator Learning for efficient Quantum Computation

Variational method with one ancilla and backpropagation approximates propagators and differential operators more efficiently than standard e

abstract click to expand

An efficient implementation of quantum algorithms is often hindered by the lack of efficient primitives for operators and state preparation. This limits both the ability of near-term quantum hardware to simulate complex problems and the potential of fault-tolerant algorithms to achieve practical quantum advantage. To address this, we propose a full-stack variational framework that transforms arbitrary operators to compact quantum circuits. The resulting variational circuits can be tailored to the connectivity and long-range interaction of the target hardware. The learning process employs backpropagation together with a cost function that efficiently optimizes unitary operators and non-unitary -- dense or sparse -- operators using only a single ancilla qubit for block encoding. Additionally, we introduce a regularization term that reduces the approximation error. The approach is validated for both quantum mechanical and engineering applications. In the former case, we learn propagators that arise in native quantum problems -- such as quantum simulation and quantum chemistry -- and achieve improved resource scaling in comparison to standard Suzuki-Trotter expansions. In the latter case, we demonstrate the approach's ability to implement the second-order central finite difference approximation of the Laplace operator -- relevant for solving partial differential equations -- while improving upon current error metrics. The final example deals with learning a dense, non-unitary operator that arises in the analysis of inviscid potential flow around an airfoil. This universality of the framework opens the door for solving general problems beyond prototypical engineering and quantum applications.

0

cs.CE 2026-06-19

Model generates 2400 CT-matched 3D fiber microstructures

by Mohamad A. Raja, Clemens Dransfeld +1 more

Autoregressive Modelling and Synthetic Generation of High-Fidelity, Statistically Equivalent 3D Microstructures for As-Manufactured Misalignments in Fiber-Reinforced Composites

Autoregressive copula framework with Bayesian tuning produces simulation-ready composites matching misalignment statistics within 10 percent

abstract click to expand

This study presents an integrated framework for processing, modelling, and generating statistically representative three-dimensional fiber microstructures from experimental X-ray-$\mu$CT observations. First, an analytical slice-segment ellipse-intersection method is introduced to extract per-slice and per-fiber in-plane and out-of-plane misalignment profiles along the fiber depth. These descriptors are then used to construct a stochastic model that captures slice-wise misalignment distributions and their depth-wise evolution through, copula-based in-plane dependence, latent autoregressive continuity, and rare extreme-misalignment motifs. The model hyperparameters are calibrated using Bayesian optimization, achieving close agreement with the original statistical descriptors, with deviations generally below 10\%. The optimized statistical model is coupled with a physical generation strategy that begins with variable-radius fiber seeding layer and proceeds through an iterative slice-by-slice 3D growth scheme, where the statistical layer guides fiber evolution and Delaunay-based neighbourhood construction with ellipse-based contact resolution ensures non-overlapping, radius-augmented synthetic microstructures. The framework successfully generates about 2400 synthetic fibers while preserving strong statistical fidelity to the original X-ray-$\mu$CT data. The proposed pipeline provides a promising and scalable route for generating statistically equivalent, geometrically admissible, and simulation-ready fiber composite microstructures for virtual testing and analysis.

0

cs.CE 2026-06-19

Orchestration gap explains stalled automation in complex industries

by Jiechao Gao, Yuandong Pan. Yuangang Li +3 more

The Orchestration Gap: Why Process Automation Stalls in Operationally Complex Industries

The needed guarantee shifts from constraint enforcement under regulatory friction to explainability under liability friction.

abstract click to expand

Agentic systems have advanced quickly on digitally native tasks, yet they have barely touched the industries where coordinated automation could matter most: logistics, healthcare operations, construction, and the many sectors whose work is spread across incompatible tools and many hands. We argue that the reason is a missing abstraction. The value in these settings does not come from a single capable model invocation; it comes from \emph{orchestration}, the runtime that coordinates multi-step workflows, enforces hard domain constraints, manages human approval, and bridges legacy systems. We develop this idea into a usable conceptual frame. We give an operational test for which workflows are orchestration-bound, a decomposition that separates how tangled a workflow is from how much of its effort is coordination and what that coordination is worth, and a feature-level account of why today's multi-agent frameworks leave a specific gap. We then advance our central claim: the right automation path is staged, and which architectural guarantee carries the most weight depends on a sector's dominant source of friction. Constraint enforcement is load-bearing under regulatory friction; explainability is load-bearing under liability friction. We close with the research program this view implies.

0