The High-Resolution Rapid Refresh (HRRR) model is a convection-allowing model used in operational weather forecasting across the contiguous United States (CONUS). To provide a computationally efficient alternative, we introduce HRRRCast, a data-driven emulator built with advanced machine learning techniques. HRRRCast includes two architectures: a ResNet-based model (ResHRRR) and a Graph Neural Network-based model (GraphHRRR). ResHRRR uses convolutional neural networks enhanced with squeeze-and-excitation blocks and Feature-wise Linear Modulation, and supports probabilistic forecasting via the Denoising Diffusion Implicit Model (DDIM). To better handle longer lead times, we train a single model to predict multiple lead times (1h, 3h, and 6h), then use a greedy rollout strategy during inference. When evaluated on composite reflectivity over the full CONUS domain using ensembles of 3 to 10 members, ResHRRR outperforms HRRR forecast at light rainfall threshold (20 dBZ) and achieves competitive performance at moderate thresholds (30 dBZ). Our work advances the StormCast model of Pathak et al. [21] by: a) training on the full CONUS domain, b) using multiple lead times to improve long-range skill, c) training on analysis data instead of the +1h post-analysis data inadvertently used in StormCast, and d) incorporating future GFS states as inputs, enabling downscaling that improves long-lead accuracy. Grid-, neighborhood-, and object-based metrics confirm better storm placement, lower frequency bias, and higher success ratios than HRRR. HRRRCast ensemble forecasts also maintain sharper spatial detail, with power spectra more closely matching HRRR analysis. While GraphHRRR underperforms in its current form, it lays groundwork for future graph-based forecasting. HRRRCast represents a step toward efficient, data-driven regional weather prediction with competitive accuracy and ensemble capability.

Parallel coordinate plots (PCPs) are a prevalent method to interpret the relationship between the control parameters and metrics. PCPs deliver such an interpretation by color gradation based on a single metric. However, it is challenging to provide such a gradation when multiple metrics are present. Although a naive approach involves calculating a single metric by linearly weighting each metric, such weighting is unclear for users. To address this problem, we first propose a principled formulation for calculating the optimal weight based on a specific preferred metric combination. Although users can simply select their preference from a two-dimensional (2D) plane for bi-metric problems, multi-metric problems require intuitive visualization to allow them to select their preference. We achieved this using various radar charts to visualize the metric trade-offs on the 2D plane reduced by UMAP. In the analysis using pedestrian flow guidance planning, our method identified unique patterns of control parameter importance for each user preference, highlighting the effectiveness of our method.

We study A/B testing, the standard protocol for measuring the performance gain of a new decision system relative to a baseline. Traditional A/B testing treats both systems as black boxes, ignoring potential similarities between them. In practice, however, new and baseline systems are rarely radically different and often share significant structure, which can be captured by their propensities to make similar decisions. We show that in such cases, the commonly used difference-in-means estimator, though unbiased, is statistically suboptimal. Leveraging off-policy estimation, we introduce a family of A/B testing estimators that exploit the propensities of the tested systems to achieve improved concentration properties. This family is flexible enough to be tailored to practical decision-making. The resulting estimators are simple, robust to propensities misspecification, substantially more accurate when the tested systems exhibit similarities, and gracefully fall back to the difference-in-means estimator when such similarities are absent. Our theoretical analysis and empirical studies confirm their efficiency and practicality.

Medical image segmentation is a fundamental task in computer-aided diagnosis and treatment. Existing approaches based on CNNs, ViTs, Mamba, and hybrid models still suffer from limitations such as restricted receptive fields, high computational cost, or insufficient accuracy. Recently, Vision Receptive-field Weighted Key-Value (VRWKV) models have emerged as a promising alternative,delivering strong long-range dependency modeling for visual tasks. However, current studies on VRWKV-based medical image segmentation mainly focus on hybrid architectures trained from scratch, while the potential of large-scale pretrained pure VRWKV models remains unexplored. In this work, we systematically investigate the effectiveness of pure VRWKV architectures for medical image segmentation. We construct Med-URWKV-T and Med-URWKV-S by reusing pretrained VRWKV encoders at different scales and pairing them with pure VRWKV decoders, enabling a comprehensive evaluation of pretrained pure VRWKV models in this domain. To further enhance performance, we propose two VRWKV-compatible modules: a Frequency-Aware Wavelet Attention (FAWA) module, which exploits wavelet transforms to capture edge details and structural characteristics, and a Multi-Scale Channel Fusion (MSCF) module, which integrates multi-scale features to strengthen informative channel representations. By incorporating them into Med-URWKV-T, we obtain the enhanced model Med-URWKV†. Extensive experiments on five medical image segmentation datasets demonstrate that Med-URWKV achieves performance comparable to or superior to state-of-the-art methods and carefully designed hybrid VRWKV architectures. Moreover, Med-URWKV† further improves segmentation accuracy, surpassing Med-URWKV-S while using only half of its parameter count, and achieves the highest average Dice similarity coefficient of 88.00%. The codes will be released.

We study parameterizations of stabilizing nonlinear policies for learning-based control. We propose a structure based on a nonlinear version of the Youla-Kucera parameterization combined with robust neural networks such as the recurrent equilibrium network (REN). The resulting parameterizations are unconstrained, and hence can be searched over with first-order optimization methods, while always ensuring closed-loop stability by construction. We study the combination of (a) nonlinear dynamics, (b) partial observation, and (c) incremental closed-loop stability requirements (contraction and Lipschitzness). We find that for the combination of (c) with either (a) or (b), a contracting and Lipschitz Youla parameter always leads to contracting and Lipschitz closed loops. However, if all three hold, then incremental stability can be lost with exogenous disturbances. Instead, a weaker condition is maintained, which we call d-tube contraction and Lipschitzness. We further obtain converse results showing that the proposed parameterization covers all contracting and Lipschitz closed loops for certain classes of nonlinear systems. Numerical experiments illustrate the utility of our parameterization when learning controllers with built-in stability certificates for: (i) ``economic'' rewards without stabilizing effects; (ii) short training horizons; and (iii) uncertain systems.

The sample covariance matrix is a cornerstone of multivariate statistics, but it is highly sensitive to outliers. These can be casewise outliers, such as cases belonging to a different population, or cellwise outliers, which are deviating cells (entries) of the data matrix. Recently some robust covariance estimators have been developed that can handle both types of outliers, but their computation is only feasible up to at most 20 dimensions. To remedy this we propose the cellRCov method, a robust covariance estimator that simultaneously handles casewise outliers, cellwise outliers, and missing data. It relies on a decomposition of the covariance on principal and orthogonal subspaces, leveraging recent work on robust PCA. It also employs a ridge-type regularization to stabilize the estimated covariance matrix. We establish some theoretical properties of cellRCov, including its casewise and cellwise influence functions as well as consistency and asymptotic normality. A simulation study demonstrates the superior performance of cellRCov in contaminated and missing data scenarios. Furthermore, its practical utility is illustrated in a real-world application to anomaly detection. We also construct and illustrate the cellRCCA method for robust and regularized canonical correlation analysis.

Respiratory viral infections pose a global health burden, yet the cellular immune mechanisms underlying protection and pathology remain unclear. Natural infection cohorts often lack pre-exposure baselines and time-controlled sampling, whereas inoculation and vaccination trials generate well-structured longitudinal transcriptomic data. However, these datasets are scattered across repositories and processed inconsistently, hindering integrative and AI-driven analyses. To address these challenges, we developed the Human Respiratory Viral Immunization LongitudinAl Gene Expression (HR-VILAGE-3K3M) repository: an AI-ready resource integrating bulk and single-cell transcriptomic profiles from 3,178 subjects across 66 studies. The dataset spans vaccination, inoculation, and mixed exposures, with samples from blood and nasal swabs collected from public repositories including GEO, ImmPort, and ArrayExpress. We curated and harmonized subject-level metadata, standardized outcome measures, and applied unified preprocessing with rigorous quality control. We further provide benchmark analyses illustrating its utility. This resource supports discovery of biomarkers, immune mechanisms, and methodological development. As one of the largest longitudinal transcriptomic resources for human respiratory viral immunization, HR-VILAGE-3K3M enables reproducible and scalable analyses to accelerate vaccine and antiviral research.

Atmospheric predictability research has long held that rapid error growth at small spatial scales imposes an intrinsic limit of roughly two weeks on deterministic weather forecast skill. We challenge this limit using GraphCast, a machine-learning weather model, by optimizing initial conditions for twice-daily forecasts spanning 2020. This approach yields an average error reduction of 86% at ten days relative to control forecasts from reanalysis initial conditions, with skill lasting beyond 30 days. Mean optimal initial-condition perturbations reveal large-scale, spatially coherent corrections primarily reflecting an intensification of the Hadley circulation. Forecasts using GraphCast-optimal initial conditions in the Pangu-Weather model achieve a 21% error reduction, peaking at four days, indicating that analysis corrections reflect adjustments that target both model and analysis error. These results demonstrate the existence of initial conditions producing skillful deterministic forecasts far beyond two weeks. Whether such initial conditions can be identified in real-time for improving operational weather forecasts remains a topic of future research.

As artificial intelligence (AI) reshapes industries and societies, ensuring its trustworthiness-through mitigating ethical risks like bias, opacity, and accountability deficits-remains a global challenge. International Organization for Standardization (ISO) AI standards, such as ISO/IEC 24027 and 24368, aim to foster responsible development by embedding fairness, transparency, and risk management into AI systems. However, their effectiveness varies across diverse regulatory landscapes, from the EU's risk-based AI Act to China's stability-focused measures and the U.S.'s fragmented state-led initiatives. This paper introduces a novel Comparative Risk-Impact Assessment Framework to evaluate how well ISO standards address ethical risks within these contexts, proposing enhancements to strengthen their global applicability. By mapping ISO standards to the EU AI Act and surveying regulatory frameworks in ten regions-including the UK, Canada, India, Japan, Singapore, South Korea, and Brazil-we establish a baseline for ethical alignment. The framework, applied to case studies in the EU, US-Colorado, and China, reveals gaps: voluntary ISO standards falter in enforcement (e.g., Colorado) and undervalue region-specific risks like privacy (China). We recommend mandatory risk audits, region-specific annexes, and a privacy-focused module to enhance ISO's adaptability. This approach not only synthesizes global trends but also offers a replicable tool for aligning standardization with ethical imperatives, fostering interoperability and trust in AI worldwide. Policymakers and standards bodies can leverage these insights to evolve AI governance, ensuring it meets diverse societal needs as the technology advances.

In this work, we initiate the study of Hamiltonian learning for positive temperature bosonic Gaussian states, the quantum generalization of the widely studied problem of learning Gaussian graphical models. We obtain efficient protocols, both in sample and computational complexity, for the task of inferring the parameters of their underlying quadratic Hamiltonian under the assumption of bounded temperature, squeezing, displacement and maximal degree of the interaction graph. Our protocol only requires heterodyne measurements, which are often experimentally feasible, and has a sample complexity that scales logarithmically with the number of modes. Furthermore, we show that it is possible to learn the underlying interaction graph in a similar setting and sample complexity. In addition, we use our techniques to obtain the first results on learning Gaussian states in trace distance with a quadratic scaling in precision and polynomial in the number of modes, albeit imposing certain restrictions on the Gaussian states. Our main technical innovations are several continuity bounds for the covariance and Hamiltonian matrix of a Gaussian state, which are of independent interest, combined with what we call the local inversion technique. In essence, the local inversion technique allows us to reliably infer the Hamiltonian of a Gaussian state by only estimating in parallel submatrices of the covariance matrix whose size scales with the desired precision, but not the number of modes. This way we bypass the need to obtain precise global estimates of the covariance matrix, controlling the sample complexity.

Spatial transcriptomics (ST) provides essential spatial context by mapping gene expression within tissue, enabling detailed study of cellular heterogeneity and tissue organization. However, aligning ST data with histology images poses challenges due to inherent spatial distortions and modality-specific variations. Existing methods largely rely on direct alignment, which often fails to capture complex cross-modal relationships. To address these limitations, we propose a novel framework that aligns gene and image features using a ranking-based alignment loss, preserving relative similarity across modalities and enabling robust multi-scale alignment. To further enhance the alignment's stability, we employ self-supervised knowledge distillation with a teacher-student network architecture, effectively mitigating disruptions from high dimensionality, sparsity, and noise in gene expression data. Extensive experiments on seven public datasets that encompass gene expression prediction, slide-level classification, and survival analysis demonstrate the efficacy of our method, showing improved alignment and predictive performance over existing methods.

Is there a statistical difference between Naive Bayes and Random Forest in terms of recall, f-measure, and precision for predicting software defects? By utilizing systematic literature review and meta-analysis, we are answering this question. We conducted a systematic literature review by establishing criteria to search and choose papers, resulting in five studies. After that, using the meta-data and forest-plots of five chosen papers, we conducted a meta-analysis to compare the two models. The results have shown that there is no significant statistical evidence that Naive Bayes perform differently from Random Forest in terms of recall, f-measure, and precision.

We develop a new approach for clustering non-spherical (i.e., arbitrary component covariances) Gaussian mixture models via a subroutine, based on the sum-of-squares method, that finds a low-dimensional separation-preserving projection of the input data. Our method gives a non-spherical analog of the classical dimension reduction, based on singular value decomposition, that, among several other applications, forms a key component of the celebrated spherical clustering algorithm of Vempala and Wang [VW04]. As applications, we obtain an algorithm to (1) cluster an arbitrary total-variation separated mixture of $k$ centered (i.e., zero-mean) Gaussians with $n\geq \operatorname{poly}(d) f(w_{\min}^{-1})$ samples and $\operatorname{poly}(n)$ time, and (2) cluster an arbitrary total-variation separated mixture of $k$ Gaussians with identical but arbitrary unknown covariance with $n \geq d^{O(\log w_{\min}^{-1})} f(w_{\min}^{-1})$ samples and $n^{O(\log w_{\min}^{-1})}$ time. Here, $w_{\min}$ is the minimum mixing weight of the input mixture, and $f$ does not depend on the dimension $d$. Our algorithms naturally extend to tolerating a dimension-independent fraction of arbitrary outliers. Before this work, the techniques in the state-of-the-art non-spherical clustering algorithms needed $d^{O(k)} f(w_{\min}^{-1})$ samples and time for clustering such mixtures. Our results may come as a surprise in the context of the $d^{Ω(k)}$ statistical query and sum-of-squares lower bounds [DKS17, DKPP24] for clustering non-spherical Gaussian mixtures. While these results are usually thought to rule out $d^{o(k)}$ cost algorithms for the problem, our results show that the lower bounds can in fact be circumvented for a remarkably general class of Gaussian mixtures.

The potential for exploitation of AI models has increased due to the rapid advancement of Artificial Intelligence (AI) and the widespread use of platforms like Model Zoo for sharing AI models. Attackers can embed malware within AI models through steganographic techniques, taking advantage of the substantial size of these models to conceal malicious data and use it for nefarious purposes, e.g. Remote Code Execution. Ensuring the security of AI models is a burgeoning area of research essential for safeguarding the multitude of organizations and users relying on AI technologies. This study leverages well-studied image few-shot learning techniques by transferring the AI models to the image field using a novel image representation. Applying few-shot learning in this field enables us to create practical models, a feat that previous works lack. Our method addresses critical limitations in state-of-the-art detection techniques that hinder their practicality. This approach reduces the required training dataset size from 40000 models to just 6. Furthermore, our methods consistently detect delicate attacks of up to 25% embedding rate and even up to 6% in some cases, while previous works were only shown to be effective for a 100%-50% embedding rate. We employ a strict evaluation strategy to ensure the trained models are generic concerning various factors. In addition, we show that our trained models successfully detect novel spread-spectrum steganography attacks, demonstrating the models' impressive robustness just by learning one type of attack. We open-source our code to support reproducibility and enhance the research in this new field.

This paper presents the results of finetuning large language models (LLMs) for the task of detecting vulnerabilities in source code. We leverage WizardCoder, a recent improvement of the state-of-the-art LLM StarCoder, and adapt it for vulnerability detection through further finetuning. To accelerate training, we modify WizardCoder's training procedure, also we investigate optimal training regimes. For the imbalanced dataset with many more negative examples than positive, we also explore different techniques to improve classification performance. The finetuned WizardCoder model achieves improvement in ROC AUC and F1 measures on balanced and imbalanced vulnerability datasets over CodeBERT-like model, demonstrating the effectiveness of adapting pretrained LLMs for vulnerability detection in source code. The key contributions are finetuning the state-of-the-art code LLM, WizardCoder, increasing its training speed without the performance harm, optimizing the training procedure and regimes, handling class imbalance, and improving performance on difficult vulnerability detection datasets. This demonstrates the potential for transfer learning by finetuning large pretrained language models for specialized source code analysis tasks.

Autism is a developmental disorder that manifests in early childhood and persists throughout life, profoundly affecting social behavior and hindering the acquisition of learning and social skills in those diagnosed. As technological advancements progress, an increasing array of technologies is being utilized to support the education of students with Autism Spectrum Disorder (ASD), aiming to improve their educational outcomes and social capabilities. Numerous studies on autism intervention have highlighted the effectiveness of social robots in behavioral treatments. However, research on the integration of social robots into classroom settings for children with autism remains sparse. This paper describes the design and implementation of a group experiment in a collective classroom setting mediated by the NAO robot. The experiment involved special education teachers and the NAO robot collaboratively conducting classroom activities, aiming to foster a dynamic learning environment through interactions among teachers, the robot, and students. Conducted in a special education school, this experiment served as a foundational study in anticipation of extended robot-assisted classroom sessions. Data from the experiment suggest that ASD students in classrooms equipped with the NAO robot exhibited notably better performance compared to those in regular classrooms. The humanoid features and body language of the NAO robot captivated the students' attention, particularly during talent shows and command tasks, where students demonstrated heightened engagement and a decrease in stereotypical repetitive behaviors and irrelevant minor movements commonly observed in regular settings. Our preliminary findings indicate that the NAO robot significantly enhances focus and classroom engagement among students with ASD, potentially improving educational performance and fostering better social behaviors.

This paper presents a method to synthesize neural network controllers to maximize reward subject to the hard constraint that the feedback system of plant and controller be dissipative, certifying requirements such as stability and $L_2$ gain bounds. It considers nonlinear and uncertain plants, modeled as the interconnection of a linear time-invariant (LTI) system and an uncertainty block, which incorporates nonlinearities. The uncertainty of the plant and the activation functions of the neural network are both described using integral quadratic constraints (IQCs). First, a dissipativity condition is derived for uncertain LTI systems. Second, this condition is used to construct a linear matrix inequality (LMI) which can be used to synthesize neural network controllers. Finally, this convex condition is used in a projection-based training method to synthesize neural network controllers with dissipativity guarantees. Numerical examples on an inverted pendulum and a flexible rod on a cart are provided to demonstrate the effectiveness of this approach.

This paper presents an autoencoder with ordered variance (AEO), in which the conventional reconstruction loss is augmented by a variance-based regularization term that promotes an ordered structure within the latent space. In this structure, the latent variables are ordered by their variance computed over the training data, facilitating systematic determination of the latent space dimensionality. The AEO is further extended using residual networks, resulting in a ResNet-based AEO (RAEO). Both AEO and RAEO green lead to discovery of nonlinear relationships among variables in unlabeled datasets, thereby enabling unsupervised static model extraction. Theoretical contributions include formal guarantees on the ordering of latent variances. The practical utility of the framework is demonstrated through its application to the identification of nonlinear steady-state models and their use in real-time optimization, with a continuous stirred tank reactor process serving as a representative case study.

In this paper we prove Gamma-convergence of a nonlocal perimeter of Minkowski type to a local anisotropic perimeter. The nonlocal model describes the regularizing effect of adversarial training in binary classifications. The energy essentially depends on the interaction between two distributions modelling likelihoods for the associated classes. We overcome typical strict regularity assumptions for the distributions by only assuming that they have bounded $BV$ densities. In the natural topology coming from compactness, we prove Gamma-convergence to a weighted perimeter with weight determined by an anisotropic function of the two densities. Despite being local, this sharp interface limit reflects classification stability with respect to adversarial perturbations. We further apply our results to deduce Gamma-convergence of the associated total variations, to study the asymptotics of adversarial training, and to prove Gamma-convergence of graph discretizations for the nonlocal perimeter.

Agent skills occupy a privileged position in the agent workflow, as agents are expected to implicitly follow and execute them, rendering third-party skills a vulnerable attack surface. Existing studies have revealed unsafe agent behaviors induced by skill-based attacks, but they primarily evaluate poisoned skills within a single task execution and enumerate harms through ad-hoc risk lists. To bridge these gaps, we introduce SkillHarm, a benchmark of skill-based attacks across the skill-use lifecycle, paired with a systematic taxonomy of skill-relevant risks. SkillHarm evaluates two attack scenarios: Fixed-Payload Poisoning (FPP), where a fixed poisoned skill package directly compromises any task session that invokes it, and Self-Mutating Poisoning (SMP), where an initially benign execution silently mutates persistent skill content, deferring harm until a subsequent reuse. It further defines 12 risk types based on the agent workflow component targeted by the harm: data pipelines, system environments, and agent autonomy. To instantiate these attacks at scale, we build AutoSkillHarm, an automated construction pipeline with coding agents driven by natural-language harnesses. The resulting benchmark contains 879 attack samples across 71 skills. Experiments show that current agents remain vulnerable with attack success rates up to 86.3% in FPP and 69.3% in SMP. Our analysis further reveals a latent risk: many apparent attack failures stem from the agent failing to engage with the poisoned file rather than genuine resistance, and current defenses still fail to reliably mitigate the threat.

Optimal transport (OT) provides a principled framework for mapping between probability distributions. Despite extensive progress, applying OT to large-scale data remains computationally demanding, and the resulting pointwise transport plans are often difficult to interpret. We introduce Optimal Mixture Transport (OMT), a scalable framework that shifts the transport paradigm from individual samples to mixtures of subpopulations, reformulating the transport problem as a strictly biconvex optimization with a unique global minimizer. We further establish theoretical guarantees on the stability of the OMT map, showing that bounded perturbations of the underlying distributions lead to bounded changes in the transport plan. By formulating subpopulations as exponential-family distributions, OMT decouples computational complexity from the sample size, scaling solely with the number of mixture components. We demonstrate the effectiveness and practicality of OMT on a wide range of synthetic benchmarks and real-world datasets, including image data and large-scale single-cell RNA sequencing measurements.

Time series forecasting has advanced rapidly, especially with the emergence of foundation models that show strong zero-shot performance on numerical extrapolation. However, in real-world forecasting settings, a statistically plausible baseline is rarely the final forecast used in practice. Before a forecast becomes decision-ready, it often needs to be revised using weakly structured business context such as holiday effects, campaign plans, external events, historical analogs, and expert feedback. This practical stage remains underexplored in the forecasting literature. In this paper, we formulate this stage as the \textbf{last-mile forecasting} problem and present an LLM-agent framework that sits on top of a forecasting backbone. Our system maintains a unified forecast workspace, invokes tools to retrieve contextual evidence, and converts reasoning trajectories into explicit forecast revision actions under structural safety constraints. It also supports long-horizon forecasting through map-reduce-style decomposition and post-hoc reflection through a memory bank. The resulting system is designed to be controllable and auditable. Through real-world case studies, we show how LLM agents can bridge the gap between statistical prediction and business-ready forecasting.

Large image datasets have accelerated progress in cognitive neuroscience and computer vision. However, most datasets are low-resolution, internet-sourced JPEGs with unknown capture conditions and limited spatial context. Places in the Wild is a dataset of 67,574 high-resolution photographs collected in situ across 810 physical locations spanning 260 basic-level scene categories, including indoor, urban, and natural environments. At each location, a 45-megapixel Canon EOS R5 mounted on a panoramic tripod captured 72 images at 5-degree horizontal intervals plus 12 images at varying elevations, yielding dense 360-degree viewpoint sampling. All images were recorded simultaneously as 14-bit RAW (CR3) files and compressed JPEGs, preserving sensor-level detail for analyses of luminance, contrast, color, and other image statistics. The dataset is accompanied by complete EXIF metadata and a suite of image-quality metrics. Places in the Wild supports research on viewpoint-dependent recognition in humans and models, training and evaluation of scene-understanding systems under realistic conditions, characterization of natural scene statistics, and experiments requiring near-full-field visual displays.

Organizations routinely run experiments for A/B testing, yet the data generated from one experiment is underutilized to inform subsequent intervention design. Significant barriers exist to extracting actionable knowledge from prior experimental data to inform new interventions. We study whether tool-augmented agentic AI can automatically learn from experimental data to generate new interventions in subsequent experiments. Through two-stage field experiments in healthcare prescription messaging (693,139 patient visits), we compare a Human + Chatbot method (Stage 1: behavioral experts with conversational AI co-designing 13 message variants, 444,691 patient visits) against a Tool-Augmented Agentic AI method (Stage 2: AI autonomously extracting principles from Stage 1 data to generate 17 new variants, 248,448 patient visits). The Agentic AI method, equipped with analytical tools, structured Data-Information-Knowledge-Wisdom (DIKW) reasoning agents, and transparent evidence chains, produces superior interventions: the best AI-generated message achieved a 69.8% CTR (+6.5 percentage points over baseline). Critically, our results suggest that the value comes from domain-specific experimental data, not from general reasoning ability: frontier LLMs operating without experimental data failed to predict which interventions would succeed. The field experiments also revealed that general-purpose behavioral theories used for intervention design do not extend uniformly to specific healthcare contexts, motivating an agentic AI approach to theory audits at field-experiment scale. Our research shows that tool-augmented AI can learn from experimental data and generate improved domain-relevant interventions, transforming behavioral experimentation from one-shot evaluation into a scalable system for cumulative design learning.

Learned heuristics have recently become a competitive alternative to traditional domain-independent heuristics for satisficing planning. Existing approaches, however, focus on improving search guidance rather than guaranteeing admissibility, which makes them unsuitable for optimal classical planning. We present the first method for learning domain-dependent heuristics that are admissible by design and thus preserve the optimality guarantees of A* search. Instead of learning a direct mapping from states to heuristic values, we learn to construct abstractions that induce admissible heuristics. We use an LLM-driven evolutionary program-synthesis framework to obtain, for each domain, a program that produces a pattern collection for any task in that domain, and we combine the resulting patterns admissibly via saturated cost partitioning. Empirically, the learned programs encode interpretable domain-specific insights, run with negligible overhead at test time and yield heuristics that match the coverage of state-of-the-art domain-independent baselines on several domains while evaluating each state substantially faster.

Progress in tabular machine learning has largely focused on increasingly sophisticated model architectures. At the same time, feature engineering remains a critical yet underexplored component of real-world modeling pipelines that is entirely absent from modern benchmarks, which creates an unquantified evaluation gap. In this work, we introduce TabPrep, a lightweight preprocessing pipeline composed of feature generators that are carefully designed to target three specific structural data patterns. We show that many widely used model classes exhibit predictable blind spots to these patterns and that systematic feature engineering alone can establish new peak performance. Across the TabArena benchmark, integrating TabPrep into model training and tuning consistently improves performance for tree-based, neural, linear, and foundation models, often surpassing gains achieved by model-centric innovations alone. TabPrep outperforms previous automated feature engineering approaches in performance, efficiency, and applicability across datasets, enabling integration into large-scale benchmarks. By releasing TabPrep (see https://github.com/atschalz/tabprep), we enable researchers to integrate feature engineering into their benchmarking setup, filling a longstanding gap in tabular evaluations.

Earth Observation (EO) has fundamentally transformed the monitoring of environmental processes and human activities up to planetary scale. Recent advances in self-supervised learning have given rise to Earth Observation Foundation Models (EOFMs), which leverage petabyte-scale unlabeled EO data to learn transferable representations across a wide range of downstream geospatial tasks. Despite these advances, current EOFMs remain largely confined to raster modalities, overlooking the rich, structured information encoded in openly-accessible vector data sources such as OpenStreetMap and Overture. Vector data provides explicit and compact representations of geographic entities, including geometry, topology, and semantic relationships, offering critical contextual signals that are often ambiguous or inaccessible in imagery alone. Raster and vector data thus represent complementary views of geographic space: raster data captures continuous physical and spectral patterns, while vector data encodes discrete objects and their relational structure and often represents more of the human rather than the physical systems (e.g. social or demographic data). However, existing geospatial representation learning paradigms treat these modalities in isolation, relying on imperfect and often lossy transformations to bridge them. This perspective paper calls for a paradigm shift toward joint Spatial Representation Learning (SRL) in an unified embedding space that integrate raster perception with vector-based reasoning. Building on emerging efforts in multimodal geospatial learning, we highlight conceptual foundations, technical challenges, and promising directions for aligning heterogeneous spatial data sources. We contend that such integration is essential for developing next-generation geospatial AI systems capable of more accurate, interpretable, and semantically grounded understanding of the Earth.

Search agents are often trained as policies over growing transcripts: the model must decide how to search while also remembering what it has seen, which evidence is useful, which constraints remain open, and which claims have actually been checked. We argue that this formulation puts too much routine state management inside the policy: reinforcement learning is forced to optimize both semantic search decisions and recoverable bookkeeping that the environment can maintain more reliably. We introduce Harness-1, a 20B search agent (retrieval subagent) trained with reinforcement learning inside a stateful search harness. The harness maintains environment-side working memory, including a candidate pool, an importance-tagged curated set, compact evidence links, verification records, compressed and deduplicated observations, and budget-aware context rendering. The policy retains the semantic decisions: what to search, which documents to keep or discard, what to verify, and when to stop. Across eight retrieval benchmarks spanning web, finance, patents, and multi-hop QA, Harness-1 achieves 0.730 average curated recall, outperforming the next strongest open search subagent by +11.4 points and remaining competitive with much larger frontier-model searchers. Its gains are especially strong on held-out transfer benchmarks, suggesting that reinforcement learning over explicit search state can produce retrieval behaviors that generalize beyond the training domains. Our code is available at https://github.com/pat-jj/harness-1.

Flapping-wing aerial vehicles (FWAVs) demonstrate remarkable agility but face substantial autonomy challenges due to their high sensitivity to aerodynamic disturbances and limited sensor payload capacity. Current simulation platforms typically rely on oversimplified laminar flow assumptions and idealized sensor models, failing to capture the complex turbulence patterns and perceptual limitations encountered in real-world operation. This simulation-to-reality discrepancy significantly impedes the development of robust autonomy systems for FWAVs. We introduce FWAV-Sim, a high-fidelity Unity-based simulation framework that integrates: (1) a composite aerodynamic model combining quasi-steady blade-element theory with bluff-body drag effects, (2) spatiotemporally correlated turbulence generation through fractal noise synthesis, and (3) realistic sensor simulation including noisy IMU measurements, LiDAR point clouds, and RGB camera feeds. Our platform enables scalable generation of synchronized datasets containing ground-truth vehicle states, aerodynamic forces, turbulent wind fields, and multi-modal sensor streams. Experimental validation demonstrates that autonomy pipelines (including both controllers and perception systems) developed in FWAV-Sim exhibit significantly improved simulation capability, thereby advancing the outstanding performance in simulation-based development for flapping-wing aerial systems.

Tool-augmented multimodal agents show strong benchmark gains, often taken as evidence that agents have learned to use tools. We argue that this interpretation can be premature: a tool-call trace alone does not show whether the tool supplied answer-critical information. We study two representative ``thinking with images'' agents, Thyme and DeepEyesV2, across real-world understanding, OCR, chart understanding, and mathematical reasoning. Each agent is compared with its Tool-Free counterpart and with a Pure-Text Reasoner trained from the same source pool without tool-calling trajectories. Tool access yields little consistent aggregate improvement, does not reliably reduce generated-token cost, and leaves only a small tool-only solved set: 93% of DeepEyesV2's tool-solved problems and 96% of Thyme's are also solved by at least one non-tool setting. Mechanism ablations further show that the full tool-use loop does not consistently outperform either the tool-call format or the returned execution result alone. In the settings we study, the analyzed agents appear to learn tool-calling patterns more reliably than tool-contributed capabilities, suggesting that evaluation should distinguish tool availability from whether tools actually expand what agents can solve.