Papers

We develop a certified, scalable approximation for high-dimensional Wasserstein distributionally robust portfolio optimization. For expected-utility maximization under order-one Wasserstein ambiguity, standard duality yields a semi-infinite convex program. For long-only portfolios with box support under the one-norm ground metric, an exact sample-specific vertex reformulation provides an exponential-size computational benchmark. We then majorize the utility by supporting hyperplanes and dualize the support subproblems, obtaining a finite hyperplane–dual formulation over compact polyhedral supports. Under the one-norm ground metric and polyhedral portfolio constraints, this formulation is a polynomial-size linear program. The uniform utility-approximation error bounds both the robust-value error and the near-optimality gap for the original robust problem. Experiments validate the certified approximation and demonstrate monthly 476-asset rebalancing and computational scalability to 1,000 assets.

This paper extends prior work on linguistic pattern based facility layout optimization by enhancing the LP Alinks framework with an explicit spatial uniformity criterion. While earlier studies demonstrated that linguistic patterns can effectively encode expert knowledge and guide agent based layout emergence, their optimization scope remained limited to cost oriented objectives. To address this gap, we introduce the Normalized Coverage Score (NCS), a scale adjusted measure of spatial evenness that complements the classical flow distance economic objective and enables systematic exploration of cost uniformity trade offs. A full factorial experiment comprising 324 conditions evaluates the combined effects of problem size, link density, virtual force scaling, and two families of membership functions on both objectives. Five way and nested three way ANOVAs reveal that structural factors (number of objects and link density) exert a dominant baseline influence on both economic performance and spatial uniformity. However, while economic cost is overwhelmingly governed by these structural characteristics, uniformity is substantially more sensitive to the linguistic pattern parameters, which control the dispersion compaction dynamics of the emerging layouts within the structural constraints. Based on these interactions, we derive a parameter-selection matrix that prescribes settings for cost minimization, uniformity maximization, or balanced compromise. Comparative analyses with Drezner's method, MDS, and non metric MDS demonstrate that the extended LP Alinks framework consistently attains superior uniformity while maintaining competitive economic performance, making it a robust and interpretable decision support tool for early stage facility layout design.

Reinforced concrete (RC) beams under impact are commonly assessed using central-impact configurations, but practical impacts may deviate from midspan and create unequal shear spans. This study investigates how impact eccentricity changes force transfer and damage development using a validated coupled smoothed particle hydrodynamics–finite element method (SPH–FEM) model. Concrete is modeled with SPH particles, while reinforcement, supports, and the impactor are modeled with FEM solid elements. After validation against central drop-weight tests, full-span eccentric-impact cases are compared with matched short-span references. The first contact-force peak changes only slightly with eccentricity, whereas the response distribution changes clearly. At the largest eccentricity, shorter-span shear reaches up to 2.23 times the central-impact value, showing shear-dominated redistribution. Absorbed energy per unit length follows the same trend in shorter-span, reaching up to 4.29 times the longer-span-side value. Matched references show that full-span eccentric beams can develop up to 18.4 kN higher local shear than symmetric short-span beams. Damage fields shift from symmetric central damage to asymmetric shorter-span-side damage with clearer fragmentation in low-strength cases. Eccentric impact should therefore be evaluated as a full-span shear-transfer and damage-asymmetry problem.

Large language models are increasingly used to read markets, assess risk, and allocate capital. However, reported results for LLM trading agents can be inflated by look-ahead leakage, optimistic execution, and risk mandates that are described but not enforced. We present OpenPM, an auditable point-in-time evaluation framework for LLM portfolio-management agents. In OpenPM, an agent manages a \$1M long-only book over the S&P 500 universe using market data at five-minute intervals. Every record visible to the agent must be available at the decision time. Natural-language risk mandates are converted into typed constraints and enforced on the executed portfolio. Each run produces audit artifacts, including a contamination certificate, a cost-sensitivity curve, and a constraint-adherence report. We also build a reference agent named the tiered allocator, where typed analysts score candidates, a constructor LLM proposes weights, and a deterministic critic guarantees feasibility. We isolate constructor behavior by capturing analyst evidence once and replaying it across constructor models. In our short-window case study, stronger constructors show modest and model-dependent gains over equal weighting on the same pool, but analyst quality matters more than constructor choice, and turnover is the main cost driver. All returns are upper bounds on a single frozen window without market impact, not validated alpha.

Autonomous agents choose actions using scores that may not reflect experimental success. We developed OPERA, an operator-residual framework for optical experiments. It represents experimental actions as optical operators and evaluates their outcomes using physically interpretable residuals. Operators specify executable changes to measurement, control or reconstruction, while residuals report departures from specified physical conditions. The agent uses both to select, combine or generate operators, and physical performance is evaluated independently against a withheld reference. Across three optical tasks, score-only feedback produced score increases without physical improvement in 23.6–39.0% of decisions, compared with 0.9–1.9% for operator-residual feedback. Operator-residual feedback increased the probability of reaching and maintaining task targets and reduced experimental budgets. Protocols selected in digital twins were transferred to three optical instruments, and repeated experiments showed a lower projection budget in structured-light reconstruction. Together, operators and residuals guide autonomous decisions using measurable physical evidence.

Degradation process (DP) modeling is widely used for remaining useful life (RUL) prediction, particularly when run-to-failure data are limited. Neural networks can present complex degradation trajectories without prescribing a fixed degradation function; however, recursive health indicator (HI) forecasting is prone to error accumulation and may fail to preserve the irreversible degradation trend over long horizons. To address these limitations, this study proposes a local-global feature mixer (LGFM) and trend-guided rollout-consistent (TG-RC) loss. The LGFM combines the original HI sequence with statistical and degradation-related features to reduce sensitivity to high-frequency noise and capture both local changes and global degradation states. The TG-RC loss supplements the conventional one-step mean squared error with a recursive multi-step rollout loss and a soft dynamic time warping alignment term based on a global trend prior. Consequently, it reduces the discrepancy between training and recursive inference, while guiding the predicted trajectory toward a consistent degradation direction. Experiments on two public bearing datasets show that the proposed framework improves long-term HI extrapolation stability and RUL prediction performance across different inspection times. The LGFM also maintains low computational complexity, while the TG-RC loss can be incorporated into various DP models to improve their long-term forecasting performance.

Whether financial news influences stock prices or simply reflects information already incorporated into them remains an open question in financial economics. The COVID-19 pandemic provides an opportunity to revisit this question, as it disrupted both news coverage and financial markets on an unprecedented scale. Existing studies have largely approached the problem through aggregate sentiment measures, leaving it unclear whether the observed relationships also hold at the level of individual firms. We study this question using 6.28 million news headlines covering 26 large United States firms between 2015 and 2025. After filtering the corpus to retain materially relevant firm-specific coverage, we construct daily stance measures and examine how their relationship with stock returns changed around the 2020 shock using panel regressions and vector autoregressions with data-driven structural breaks. Our findings indicate that the relationship between financial news and equity markets is more nuanced than aggregate analyses alone suggest. While we find little evidence of a persistent market-wide change in media stance or stock returns following the pandemic, dynamic relationships emerge for a subset of firms around their own structural breaks. Taken together, these results suggest that understanding media-market interactions requires firm specific analysis and provide a framework for studying how news and prices interact under changing market conditions.

Shell-based cohesive elements tend to overestimate the compression ahead of the crack tip because of approximations in the penalty stiffness. In this study, the higher-order structural cohesive element previously developed by the authors is enhanced with a resin-rich layer-dependent penalty stiffness to improve both computational efficiency and predictive accuracy. The proposed formulation distinguishes between the normal and shear penalty stiffnesses. It extends the resin-rich layer-based penalty stiffness from the layer-wise to the equivalent single-layer framework. This extension is achieved using the through-thickness distributions of the out-of-plane normal and transverse shear stresses derived from beam theory. The proposed method is verified and validated against benchmark problems for Mode I, Mode II, mixed-mode, and reinforced DCB configurations. Compared with the conventional formulation, it exhibits significantly improved mesh convergence and provides more accurate predictions of the compression distribution ahead of the crack tip and the delamination propagation.

The development of nanotherapeutics often involves extensive empirical optimization due to the sensitivity of nanoparticle properties, such as size and polydispersity index (PDI), to minor changes in process parameters. Factors like formulation concentration, flow rates, and mixing ratios can significantly influence clinical efficacy and therapeutic outcomes. The absence of predictive mathematical frameworks has made iterative experimental screening necessary, increasing both costs and development time. This study introduces and validates a predictive modeling approach based on shape constraints, aiming to enhance the estimation of nanoparticle characteristics across various process conditions. Using controlled microfluidic methods, liposomes and lipid nanoparticles were systematically prepared under varying lipid concentrations, flow rates, and aqueous-to-organic mixing ratios. The shape-constrained model, informed by both experimental data and expert knowledge, was subsequently validated for a pharmaceutical application using minimal empirical data. Results reveal that shape-constrained modeling facilitates accurate prediction of nanoparticle size and dispersity, reducing the need for extensive experimental workflows. This framework supports rational and efficient process development for manufacturing nanomedicine systems.

A novel mechanical model and corresponding finite element method for anisotropic, hyperelastic, curved membranes are proposed. Hyperelastic fibers are embedded into the otherwise isotropic membrane, being relevant, for example, in reduced models for biological tissues and textiles. The geometrically nonlinear mechanics is formulated based on first principles of continuum mechanics (finite strain theory). The employed differential operators are formulated in a coordinate-free manner, through a framework known as tangential differential calculus. This enables a (semi-)implicit description of the fiber geometry through the intersection of level sets of some scalar function with the explicitly defined membrane surface. The mechanical model of the implicit fibers is then coupled to the mechanics of the membrane. For the numerical analysis, finite elements are applied such that the resulting scheme is a hybrid between classical Surface FEM and fictitious domain methods. For smooth physical fields, higher-order convergence rates are obtained and confirm the success of the numerical method.

Dual-polarized (DP) microstrip antennas are utilized in wireless systems for efficient data transmission. However, their bandwidth is typically very limited. In this contribution, we propose to design DP microstrip antennas with enhanced bandwidth using a density-based topology optimization approach. We formulate an optimization problem that simultaneously accounts for feeding port matching, the ports' isolation, and far-field dual-polarized performance. To enhance the bandwidth, we employ an FR4 stack-up, in which the copper on two layers is optimized simultaneously. We present two antenna designs operating around 5.7 GHz, which show a compromise in performance between a high isolation (more than 40 dB) and enhanced impedance bandwidth (around 10%). The optimized designs are experimentally validated, showing an excellent agreement between the simulated and measured performance.

Scientific machine learning commonly validates models at the level of a subdomain, a benchmark split, or an explanation for one prediction. Yet such local checks cannot establish whether the resulting explanations can be assembled into one globally admissible explanation. We introduce Scientific Explanation-Admissibility Machines (SEAM), a generator-agnostic framework that makes this local-to-global consistency question computable across regions, sensors, regimes, and model components. The finite explanation-sheaf instantiation SEAM-$\Omega$ represents each region by a structured explanation with state, closure, and observation channels together with optional contract metadata; compares neighboring explanations on their overlaps; and converts disagreement into a channel-resolved obstruction. This obstruction locates inconsistency and tests competing declared accounts by restricting each repair to the revisions that one account permits. Exact feasibility refutes or retains an account; when exact repair is unavailable, residual-aware regularized records provide a separately labeled empirical attribution. The framework also separates inconsistency from non-identifiability and monitors learned generators under distribution shift. We establish theorems for minimum-cost intervention and conservation-contract detectability, together with companion results for identifiability and closure recoverability. Across nineteen experiments involving synthetic partial differential equation systems and out-of-distribution Fourier neural operator (FNO) monitoring, SEAM detects incompatible explanations even when local predictions are accurate, and attributes failures to specific channels and overlaps. SEAM adds a global explanation-consistency audit to existing solvers and learning models, testing whether their local explanations form a coherent scientific account.

This study provides a method to quantify the benefits of shifting passenger traffic from air to high-speed rail from the perspective of flight-delay cost reduction. We first estimate the number of flight reductions for airport origin-destination pairs based on the high-speed rail ridership forecasts provided in the California High-Speed Rail 2020 Business Plan, and then distribute these flight reductions to quarter-hour intervals. Lasso models are applied to estimate the impact of reduced queuing delays at SFO, LAX, and SAN on arrival delays at the national Core 29 airports. These delay reductions are then monetized using aircraft operating costs and the value of passenger time. We evaluate alternative airport-capacity and flight-schedule scenarios, as well as multiple percentiles of probabilistic high-speed rail ridership forecasts. The resulting estimates indicate flight-delay cost savings of $51-88 million in 2018 dollars in 2029 and $235-392 million in 2018 dollars in 2033.

A hybrid s-version of isogeometric analysis (hS-IGA) strategy is proposed for accurate and efficient evaluation of near-crack fracture quantities in dynamic crack propagation analysis. The strategy retains the global-local superposition framework of the conventional s-method, while introducing B-spline basis functions only into the global discretisation and preserving a Lagrange-based local mesh in the crack domain. This hybrid formulation is motivated by the continuity-related bottleneck in the global-local coupling integration of the conventional Lagrange-based s-method, and by the need to retain a Lagrange-based local mesh for crack representation and post-processing of the dynamic stress intensity factor (DSIF) and local stress. The resulting formulation removes discontinuities in the coupling integrands caused by the global approximation and enables accurate coupling integration by standard Gauss quadrature without recursive subdivision. The proposed strategy is verified using two-dimensional stationary and dynamic straight-crack problems against the standard finite element method and the conventional s-method, and is further assessed using three-dimensional stationary and dynamically propagating circular-crack problems against the conventional s-method. Results show that the proposed hS-IGA strategy accurately evaluates the DSIF and local stress while retaining the global-local modelling advantages of the s-method. It also substantially reduces the number of integration points required for coupling integration, by approximately 81% in the two-dimensional dynamic benchmark and 95.6% in the three-dimensional dynamic benchmark relative to the conventional s-method. These results demonstrate that the proposed hS-IGA framework provides an accurate and efficient global-local strategy for dynamic crack propagation analyses requiring reliable evaluation of near-crack fracture quantities.

Efficient solution of time-dependent parametric partial differential equations (PDEs) is central to computational science and engineering. Existing deep-learning-based accelerators span a physics-to-data spectrum, from physics-informed solvers with strong physical consistency but high computational cost to data-driven surrogates with efficient inference but strong dependence on high-fidelity datasets. These methods have largely evolved in isolation and are often connected only through explicit solution-field labels. We propose PHD-SF, a compact physics-to-data spectrum framework for accelerating time-dependent parametric PDEs. By combining an SD-TPD separation strategy with differential-operator learning, PHD-SF allows reusable model information, including spatial features, latent dynamical features, and spatial differential operators, to be generated, inherited, and enriched across three operating modes, PIDON, HIDON, and DIDON, within a unified DON-based architecture. This design enables label-free spectrum modeling without precomputed full-field solution labels or explicit label transfer among modes. PHD-SF further establishes a separated-solving, hyper-reduction-like enrichment, and direct-inference acceleration path, reducing the dependence on large-scale high-fidelity data generation and its offline cost. Results on four benchmark time-dependent PDEs show accurate cross-parameter solution and direct inference using only one or two representative parameter cases in the initial physics-informed stage. The total end-to-end cost is lower than that required to train a conventional PINN for a single parameter case, while supporting reusable cross-parameter inference. PHD-SF therefore provides an efficient solution path for many-query parametric PDEs.

Workforce planning is a recurring operational decision during construction projects that requires accurate forecasts of the future workforce demand for individual tasks. However, in practice, tasks have different completion dates, resulting in variable forecast horizons. In addition, the sum of the predicted daily workforce demands must equal the total workforce allocation specified in advance. Most existing machine-learning (ML)-based forecasting models assume fixed-length outputs and do not explicitly impose an aggregate demand constraint, making them unsuitable for these operational requirements. To address this problem, this study proposes constraint-preserving residual allocation forecasting (CP-RAF). The CP-RAF represents an observed workforce demand time series as a coefficient vector and retrieves completed tasks with similar temporal shapes. Then, it estimates the allocation profile over the remaining task duration using similarity-weight averaging. The predefined remaining workforce demand for each task was distributed according to the estimated profile, and the forecast horizon was adjusted while retaining profile characteristics. This procedure accommodates variable forecast horizons while preserving the aggregate demand constraints. CP-RAF was evaluated using workforce demand field data. The results showed that CP-RAF outperformed eight baseline models in medium- and long-horizon fixed-length forecasting and maintained low forecast errors under variable-length forecasting. By directly incorporating operational constraints into the forecasting procedure, the proposed method provides a framework suitable for workforce allocation in construction practices.

An AI scientist should not grade its own homework. Yet in the systems we examined, the agent that reviews the work usually comes from the same model family as the agent that produced it, or at least from the same vendor. Model evaluators are known to favour their own generations. Whether models trained alike also share blind spots is a conjecture, not a settled finding, but if they do, the reviewer inherits the author's. The record of what was flagged and what was waved through often sits in platform logs that nobody outside can replay. We present CrossAudit, a protocol for supervising autonomous research pipelines. It rests on three commitments. Each increment of work is audited by an agent from a different vendor against a rulebook a human wrote and versioned. Reports, verdicts, disputes and rulings are git commits, so the supervision history can be re-read and cited; raw model exchanges are not yet part of that record. Scripted checks run before any model does. Advisory judgement never gates the pipeline: a model blocks only by citing a rule, and no model may waive a deterministic failure. Blockers that survive a bounded number of revision rounds go to a person. We state the protocol as eight invariants. We describe a reference implementation built from GitHub Actions and a few hundred lines of Python, and report a live deployment of a closely related variant in a computational-chemistry pipeline. We also ran a seeded-defect trial (30 increments, 43 seeded defects, one run per configuration). A cross-vendor audit of our own repository then voided its blinding. We adopt that audit's findings and report the corrected results. The trial shows that two vendors read the same rulebook differently. It does not show that either is better. The strongest evidence here is the committed, uncontrolled record of cross-vendor audits of this paper itself.

Connected aftermarket devices extend vehicle diagnostics, repair workflows, and over-the-air software maintenance beyond original equipment manufacturer boundaries, yet their heterogeneous ownership and long service life complicate conventional perimeter security. This paper develops Zero Trust Federated Learning for Connected Aftermarket Devices (ZT FL CADE), an edge-learning architecture that combines device-level access control, privacy-preserving federated learning, and adversarial validation for over-the-air update and predictive maintenance decisions. The evaluation uses a single synthetic dataset of 144,000 telemetry windows from 240 devices, 12 vendor domains, and 180 days of operation. Features include bus entropy, update latency, attestation age, signature retries, environmental signals, fault-code rates, packet loss, drift, mileage, and trust score, with targets for maintenance risk, update intrusion, and access action. ZT FL CADE trains local temporal models, aggregates privacy-bounded updates, scores each device against behavioral and update integrity evidence, and routes update requests to allow, challenge, or quarantine actions. Synthetic experiments improve maintenance risk F1 from 0.837 for Fed Avg to 0.883, improve intrusion F1 from 0.856 to 0.897, and preserve 0.842 intrusion recall when 20 percent of selected clients are adversarial. Mean access-decision latency remains 44 MS, below the 100 MS operational budget used in the simulation. The results do not establish field validation, but they indicate that zero-trust policy enforcement and federated learning can be evaluated jointly rather than as separate aftermarket security controls. Index Terms Zero trust architecture, federated learning, connected aftermarket devices, over-the-air updates, adversarial machine learning, edge artificial intelligence, predictive maintenance, automotive cybersecurity

Supervised training of finite-element (FE) surrogate models requires reference solutions, and each reference solution is obtained by solving the system that the surrogate is intended to replace. The assembled discrete potential energy provides a training signal that requires no reference solution. This note records, with proofs, the identities that make this signal exact for linear elastostatics: the difference between the energy of a prediction and the energy of the reference solution equals one half of the squared stiffness-norm error, and the gradient of the energy equals the stiffness-weighted error. Label-free discrete-energy minimisation and supervised regression in the stiffness norm therefore have the same unique minimiser and identical gradients at every point. Around this central result, the note states a conditioning lemma that bounds the displacement error by the energy gap, a modewise contraction identity that explains why the Euclidean displacement error is an unsuitable primary metric, the Chebyshev bound that governs conjugate-gradient post-processing of surrogate predictions, and a conditional latent-separation proposition for joint-embedding predictive architecture (JEPA) pretraining on a shared stiffness operator, with an explicit numerical counterexample that delimits its scope. Every claim with numeric content is implemented as an executable falsification check; the checks were executed twice, on synthetic test problems and on a probe set of 16 instances from the validation split of a pre-registered experimental run, and every inequality holds, with the measured tightness reported. A closing section explains why the construction does not extend to elastodynamics through direct minimisation of the action functional, and which time-discrete formulation restores exactness.

Kinetic model discovery is a central challenge in chemical engineering, as accurate rate expressions are essential for understanding and controlling chemical and biological processes. Symbolic regression (SR) has emerged as a powerful data-driven approach for identifying interpretable kinetic models, but usually operates without domain knowledge, often exploring physicochemically implausible models. Large language models (LLMs) offer a promising avenue for injecting domain expertise into this search. Here, we introduce an LLM-guided SR framework, embedding an LLM module within an iterative SR algorithm for automated kinetic model discovery. The LLM performs two roles at each iteration: (1) a qualitative physicochemical critique of the best SR candidates, and (2) the proposal of new candidate rate expressions guided by the SR-generated models and embedded chemical knowledge. Our framework is evaluated on four in silico case studies of increasing complexity, spanning heterogeneous catalysis and bioprocess systems. Results show the LLM-guided framework reduces iterations to identify the ground-truth model by $41.7-79.3\%$ versus a state-of-the-art SR framework, with the LLM directly proposing the correct model structure in over half of the guided runs. In practical settings, where each iteration typically requires a new wet-lab experiment, this translates into a substantial reduction in experimental effort. Predictive performance on an independent validation set is equivalent between both approaches, with $R^2>0.98$ in all case studies. Ablation studies indicate that both the SR component and the LLM scale contribute to this performance, with a reduced-size LLM largely retaining discovery efficiency. These findings demonstrate that LLMs can effectively inject domain knowledge into scientific model discovery, paving the way toward fully automated, domain-aware kinetic modelling pipelines.

Projection-based reduced-order modeling (ROM) has become a cornerstone technique for accelerating parameter-intensive engineering simulations, yet the methodology remains challenging to teach and prototype. The difficulty stems from its multi-stage workflow, which encompasses snapshot generation, SVD/POD basis construction, Galerkin projection, offline-online decomposition, hyper-reduction, and error assessment. Existing software frameworks either abstract these stages behind high-level interfaces or depend on compiled, monolithic solver stacks that are difficult for students and early-stage researchers to inspect or modify. This paper introduces scikit-rom, an open-source Python library designed to make the complete projection-based ROM pipeline transparent, modifiable, and suitable for interactive notebook-based exploration. Built on the lightweight finite element backend scikit-fem, scikit-rom provides modular problem templates and dedicated facilities for snapshot generation, reduced-basis construction, reduced-operator assembly, online ROM solution, hyper-reduction, and quantitative accuracy assessment within a unified workflow. Hyper-reduction is supported through four strategies: the Discrete Empirical Interpolation Method (DEIM), S-OPT sampling, the Energy-Conserving Sampling and Weighting (ECSW) scheme, and ECM-style cubature construction. Four progressive worked examples guide the reader from full-order simulation to ROM and hyper-reduced model construction across linear, nonlinear, static, and transient problem classes. The library is intended for graduate-level instruction, intensive workshops, and research prototyping in the reduced-order modeling community. Source code and examples are available at github.com/suparnob100/scikit-rom, with documentation at scikitrom.github.io.

Solving PDE-governed physical problems is computationally expensive, limiting the availability of high-fidelity (HF) data for training neural operators, which typically require large datasets. Multi-fidelity learning addresses this by combining abundant low-fidelity (LF) data with limited HF samples. To evaluate several multi-fidelity neural operator strategies, including two-step, residual, intermediate, and transfer learning, we conduct a systematic assessment across four PDE test cases: two standard grid-based problems, a modified Darcy flow with LF-HF governing equation discrepancies, and an unsteady smoke inflow with temporal dynamics. Most prior studies focus on grid-based LF-HF differences, but real-world discrepancies, such as those between RANS and LES models in CFD, can be more complex. To capture these more challenging scenarios, we introduce two new test cases specifically designed to emulate nontrivial LF-HF mismatch beyond simple grid-resolution differences, enabling a more realistic assessment of multi-fidelity strategies under substantial model discrepancies. We find that methods directly feeding LF predictions into HF models are prone to error amplification under large LF-HF discrepancies, particularly in autoregressive settings. In contrast, transfer learning, which uses LF-trained weights to initialise the HF model, consistently improves accuracy by providing a robust prior refined with HF data. These results highlight the importance of establishing transfer learning as a reliable approach for efficient multi-fidelity PDE surrogate modelling.

Harmonized System (HS) code prediction (HSP) from commodity text is essential to international trade, and its importance continues to grow in port logistics. Recently, large language models (LLMs) have been actively investigated for this task, owing especially to their strong language-understanding capabilities. However, their high computational cost limits deployment in constrained environments such as container terminals. Small language models (SLMs) offer a practical alternative, but their smaller scale makes them prone to generating invalid HS codes and to overlooking the hierarchical semantics between commodity text and HS codes. To address these limitations, this study proposes TRIE-HSA, which combines trie-constrained token prediction with hierarchy-aware semantic alignment (HSA). This framework constrains the SLM to predict only valid digits under the HS taxonomy and aligns commodity text representations with the hierarchical structure of HS codes. In extensive experiments on data collected from an operational container terminal, TRIE-HSA improved average HS6 accuracy by 49.96% over zero-shot inference and exceeded the strongest task-specific benchmark by 11.94%. These results demonstrate that accurate and structurally valid HSP is achievable with fewer than 10 billion parameters. Therefore, TRIE-HSA offers a practical basis for deployment of HSP in port logistics operations that cannot support large scale LLMs.

Large language models are increasingly used for information seeking, yet semantically equivalent questions phrased in different ways can receive answers of considerably different quality. System prompts are widely employed to steer response behavior, but they are typically optimized for average-case quality, so some question phrasings may still receive incomplete or low-quality answers. To address this, we formulate a constrained mixed-strategy GroupDRO framework for system-prompt selection. Instead of optimizing the system-prompt text, the framework assigns weights to system prompts in an existing pool to minimize the worst-case information-quality loss across evaluation metrics and groups, while constraining the mean loss to stay close to that of average-based selection. Because pool generation and selection are decoupled, the method applies to any system-prompt pool and can leverage an ensemble of complementary system prompts rather than a single one. Across five LLMs on two bilingual medical and consumer-finance benchmarks, the constrained method reduces the Overall Mean, Worst 25% Mean, and Worst by 13.1%, 13.2%, and 13.7% on average relative to no mitigation while keeping overall quality close to Average selection. Its multi-prompt weights reveal complementarity across metric-group pairs. Code and data are available at https://github.com/Rainxu09/equitable-system-prompt-selection.

Designing multilayered materials for controlled deformation is heavily bottlenecked by the immense computational costs and time of traditional, industrial-grade finite element methods (FEM). This excessive expense severely limits design space exploration and gradient-based optimization. To streamline the workflow, we propose a framework for parameterized physics-informed neural networks (P2INNs). The framework encompasses three core components: I. a FEM baseline with a Hex8 element and trilinear basis functions, II. a P2INN displacement field model across variable material stiffness and layer thicknesses, and III. FEM-referenced evaluation. The PINN enforces static linear elasticity via Navier-Cauchy residuals with training strategies that prioritize physics, including layerwise PDE decomposition, interface continuity penalty, and compliance aware scaling. For one-layer, pure physics-driven configurations, the PINN achieves a mean volume MAE of 1.56% and worst-case volume MAE of 2.87% against the FEM benchmark. For three-layer configurations with controlled supervised training, the PINN achieves a mean volume MAE of 2.53% and worst-case volume MAE of 4.66%, meeting the near-5% worst-case volume-MAE target for preliminary design-space exploration. The trained P2INN model remains a lightweight model with a simple forward pass for future calculations, creating an alternative to traditional FEM. By drastically reducing the need for expensive FEM evaluations, this approach could accelerate the inverse design and optimization of advanced layered architectures for protective structures in various load-heavy or potentially collision-heavy fields.

Characterizing materials through manual extraction of physical properties from microstructure images is a laborious process. This work presents a workflow to extract porosity, solid fraction, grain size distribution, and pore size distribution from scanning electron microscopy (SEM) images of sintered ceramic samples using an automated pipeline. The primary challenge for extracting physical properties from SEM images is the presence of unimodal histograms in SEM images as a result of the overlapping intensity ranges for the grain and pore phases. We evaluated several different methods for noise reduction and local thresholding of SEM images. We find that topological filtering in combination with Sauvola thresholding enables segmentation and extraction of physical property data from SEM images. We validated the automated pipeline by comparing our results with the results of manual analyses performed for samples sintered at 1200$^o$C and 1400$^o$C and achieved an Intersection over Union (IoU) score of 95.14% and 99.85%, respectively. The workflow provides an efficient means to automatically extract microstructure properties from SEM images as a crucial step in generating materials datasets for machine learning.

We introduce Simulation-Based Imaging (SBI), a framework for non-destructive acoustic imaging in which machine learning models trained entirely on simulated data serve as real-time solvers for the acoustic inverse problem. A high-fidelity nodal Discontinuous Galerkin forward solver generates large training datasets by randomizing inclusion geometry within a unit-cube domain; a 2D convolutional neural network then learns a direct mapping from boundary pressure measurements to a 32 by 32 by 32 voxel reconstruction of the interior. The trained model reliably recovers inclusion position and size from 144 boundary sensors with no prior knowledge of inclusion count or geometry. Reconstruction error degrades by only 13% under 5% additive measurement noise, and just 17% of the sensor array (24 of 144 sensors) suffices for quality within 4% of full coverage. These results establish SBI as a viable proof-of-concept imaging device whose complexity resides in software rather than hardware, opening a path toward cheap, portable, deployable imaging systems.

Land use models often represent decision-makers as homogeneous and rational, overlooking socio-psychological diversity and potentially generating rapid, coordinated land use responses that contrast with observed land use patterns. Here, we integrate an empirical typology of European forestry practitioners into an agent-based land use model by translating survey-based behavioural profiles into cognitive parameters governing endogenous decision-making processes. We distinguish five practitioner types and parametrise heterogeneous agent populations according to the empirically observed distribution of these types. Using a stylised model landscape, we compare these heterogeneous populations against a homogeneous rational-choice baseline and single-type populations. Compared with the homogeneous rational-choice baseline, heterogeneous populations dampen synchronised responses to changing ecosystem service demand, producing more gradual land use dynamics and a higher prevalence of medium intensity management, that better reflect empirical land use patterns. Our experiments reveal that similar land use patterns can emerge through different behavioural mechanisms, while the behaviour of individual decision-making types depends strongly on the population context in which they are embedded. Together, these two findings show that individual behavioural responses and land use outcomes co-evolve rather than decision-making types mapping onto fixed management practices. Explicitly representing socio-psychological diversity can therefore improve the realism and policy-relevance of agent-based land-use models by capturing feedbacks between cognition, social interactions, and emergent land use dynamics. Empirical data may make this representation possible, but models can also use this capability to explore behavioural heterogeneity as a key source of uncertainty when data are absent.

Mean-reverting dynamics are pervasive in finance, and the Cox–Ingersoll–Ross (CIR) process is a standard model for the time series they produce, from short rates to credit default swap (CDS) spreads. Yet CIR models capture only correlated co-movement, not causal influence between series, so they cannot answer the system's response when one series is externally shocked, which observational conditionals confound with historical co-movement. We make two contributions. First, an amortized model for distributional causal effect estimation that frames trajectories as time-stamped observations and predicts the calibrated multi-horizon shock response without retraining per scenario. Second, a causal multivariate CIR data-generating process that supplies the paired observational and interventional ground truth that real markets cannot. We instantiate and calibrate the framework on CDS spreads as a testbed. CIR-ACTIVA's validity is established on synthetic ground truth, independent of how well the simulator matches reality, while practical grounding is assessed by backtesting the generated traces against real CDS data. Against observational and amortized causal-inference baselines, CIR-ACTIVA leads on both causal selectivity in the joint distribution and horizon-resolved calibration, retaining its selectivity once the interventional law varies over the horizon, with gains concentrating at short horizons. This opens up a class of what-if queries on coupled spread systems, CDS stress testing among them, that observational forecasters cannot answer.

Artificial intelligence (AI) is rapidly transforming high-skilled domains, requiring higher education institutions (HEI) to balance the teaching of foundational principles with the integration of emerging tools to ensure workforce readiness. While HEI are increasingly adopting AI, many continue to grapple with how it should be incorporated into curricula and governed through policy, especially when such policies are set at different levels of an institution. This research analyzes AI policies across HEI from 34 states in the United States to investigate what these policies entail and how policies set across institutions as well as within different levels at an institution differ. Using natural language processing (NLP) to analyze institutional AI policies, we find a clear divergence: university-level policies emphasize data security and risk mitigation whereas school-level policies, when present, focus on pedagogical applications and tool usage. When focusing on business school specific policies, relatively few business schools maintain AI policies distinct from university frameworks, creating misalignment with discipline-specific learning objectives. This gap poses challenges particularly for faculty and students as well as for accreditation purposes. Our insights suggest that guidelines should be aligned with broader institutional policies while addressing discipline-specific learning objectives and evolving workforce demands.