arXiv:2605.21143v2 Announce Type: replace
Abstract: A soundscape is composed of three types of sound: biophony (sounds made by animals), geophony (natural abiotic sounds) and anthropophony (sounds made by humans). A key research question in the field of soundscape ecology is how these components interact with each other, specifically how biophony responds to geophony and anthropophony. Nevertheless, as of today, there are not many analytical instruments that enable the distinct quantification of these elements. Recent machine learning (ML) approaches aim to support automated analysis but often rely on task-specific or clean data, limiting generalisation to noisy passive acoustic monitoring (PAM) recordings. This study presents a clear and reproducible structure to build ML models for coarse soundscape classification and introduces CoarseSoundNet, a deep learning model trained to distinguish biophony, geophony, and anthropophony under realistic PAM conditions. We systematically investigate model architectures, the influence of an additional training class, data composition, and evaluation strategies. Our findings suggest that model performance improves with additional PAM data, especially when similar to the target domain, and by introducing an explicit silence class during training. Class-specific decision thresholds and duration-based constraints further enhance performance, particularly for anthropophony and geophony. Error analyses exhibit challenges for anthropophony due to masking effects and confusions for silence and insect sounds for geophony and biophony. Finally, we conduct an ecological case study which shows that pre-filtering recordings with CoarseSoundNet yields acoustic index trends comparable to ground-truth filtering, supporting its use as an effective preprocessing tool for ecoacoustic analyses.
Science Journals
arXiv:2602.04172v4 Announce Type: replace
Abstract: Density-functional theory (DFT) has become the workhorse of modern computational chemistry, with dispersion corrections such as the exchange-hole dipole moment (XDM) model playing a key role in high-accuracy modelling of large-scale systems. All previous production implementations of XDM have used the two-parameter Becke--Johnson damping function based on atomic radii. Here, we introduce and implement a new XDM variant that uses a one-parameter damping function based on atomic numbers, recently proposed by Becke. Both this new Z damping and the canonical BJ-damping variants of XDM are benchmarked on the comprehensive GMTKN55 database using minimally empirical generalised-gradient-approximation, global hybrid, and range-separated hybrid functionals. This marks the first time that the XDM (and many-body dispersion, MBD) corrections have been tested on the GMTKN55 set. Using the new WTMAD-4 metric, an outlier analysis is performed for all new data, as well as for top-ranking functionals from the literature at each rung, providing insight into both performance and consistency across the dataset. We also extended our analysis to the DM21 and Skala machine-learned functionals that have garnered recent attention. To test Z damping's transferability to the solid state, four benchmarks involving molecular crystals are also considered. Across these molecular and solid-state benchmarks, the revPBE0 and B86bPBE0 hybrid functionals, paired with the Z-damped XDM variant, show excellent performance.
arXiv:2605.21897v1 Announce Type: new
Abstract: The highly dynamic nature of vehicular networks necessitates proactive and site-specific radio resource management (RRM) to achieve ultra-reliable low-latency communications. While Network Digital Twins (NDTs) have emerged as a promising enabler, ray-tracing remains time-consuming, challenging accurate RRM under latency constraints. We propose AdaPTwin, an adaptive multi-fidelity predictive NDT for proactive and latency-aware RRM in vehicular networks. Unlike single- and multi-fidelity NDTs with fixed fidelity levels, AdaPTwin dynamically adjusts NDT fidelity based on network conditions. The framework adopts a hierarchical cloud-edge architecture, where computationally intensive fidelity selection is performed periodically in the cloud, and the proactive RRM loop operates in real-time at the edge. The edge-based proactive RRM task consists of channel prediction between vehicles and roadside units (RSUs) via trajectory forecasting and look-ahead ray tracing, followed by RRM execution. A transformer model enhanced with continual and transfer learning enables vehicular trajectory prediction while adapting to new environments and traffic patterns. Ray-tracing is performed using NVIDIA Sionna by exploiting a dynamically updated virtual environment to ensure realistic radio propagation within the NDT. Furthermore, a joint RSU beamforming and vehicle-RSU association problem is formulated to maximize proportionally fair sum-rate, and it is efficiently solved using a scalable multi-start iterative coordinate descent algorithm. Comparisons against reactive, single-fidelity, and non-adaptive predictive NDTs under realistic vehicular conditions confirm that AdaPTwin successfully adapts to diverse scenarios where other frameworks fail. Ultimately, AdaPTwin achieves up to 90% sum-rate gain and 80% outage probability reduction compared to non-adaptive NDTs, while maintaining real-time performance.
arXiv:2605.22051v1 Announce Type: new
Abstract: Generating high-fidelity visual effects (VFX) typically demands massive datasets and prohibitive computational power due to the intricate coupling of spatial textures and temporal dynamics. In this paper, we introduce EasyVFX, a resource-efficient framework that achieves realistic VFX synthesis under stringent constraints. Our core philosophy lies in frequency-domain decomposition: we observe that the complexity of VFX can be significantly mitigated by decoupling high-frequency components, which represent intricate spatial appearances, from low-frequency components that encapsulate global motion dynamics. This spectral disentanglement transforms a high-dimensional learning problem into manageable sub-tasks, thereby lowering the optimization barrier and reducing data dependency. Building upon this insight, we propose a two-stage training paradigm. First, we design a Frequency-aware Mixture-of-Experts (Freq-MoE) architecture. By utilizing a soft routing mechanism, our model assigns specialized experts to distinct spectral bands, enabling them to cultivate robust priors for appearance and motion dynamics. This specialization allows the model to acquire foundational VFX knowledge with fewer GPU resources. Second, we introduce a Test-Time Training strategy powered by a novel Frequency-constraint Loss. This allows the pre-trained model to swiftly adapt to specific, unseen effects through localized optimizations, requiring only about 100 steps on a single GPU. Experimental results demonstrate that EasyVFX produces structurally consistent and visually stunning effects, proving that frequency-aware learning is a key catalyst for democratizing professional-grade VFX.
arXiv:2605.21491v1 Announce Type: new
Abstract: As language models accelerate scientific research by automating hypothesis generation and implementation, a new bottleneck emerges: evaluating and filtering hundreds of AI-generated ideas without exhaustive experimentation. We ask whether LMs can learn to forecast the empirical success of research ideas before any experiments are run. We study comparative empirical forecasting: given a benchmark-specific research goal and two candidate ideas, predict which will achieve better benchmark performance. We construct a dataset of 11,488 idea pairs grounded in objective outcomes from PapersWithCode. While off-the-shelf 8B-parameter models struggle (30% acc.), SFT dramatically boosts performance to 77.1%, outperforming GPT-5 (61.1%). By framing evaluation as a reasoning task via Reinforcement Learning with Verifiable Rewards (RLVR), we train models to discover latent reasoning paths, achieving 71.35% acc. with interpretable justifications. Through additional ablations and out-of-distribution tests, we show robustness to surface-level heuristics and transfer to both a cross-domain time-split test set and an independently constructed test set. Our results demonstrate that compute-efficient small language models can serve as effective, objective verifiers, offering a scalable path for autonomous scientific discovery.
arXiv:2605.21498v1 Announce Type: new
Abstract: ECDSA signatures form the bedrock of blockchain transaction authentication, yet their security critically depends on proper nonce generation. We uncover a critical vulnerability in the Polygon MEV ecosystem: systematic nonce reuse that enables complete private key recovery. Analyzing on-chain data reveals that searchers, driven by the need for sub-second response times in sealed-bid auctions, employ predictable nonce patterns. These patterns create linear relationships between signatures, allowing passive attackers to recover private keys using elementary algebra. We provide a compact linear-system formulation for such attacks, including the dangerous case of cross-wallet nonce collisions, and present concrete evidence of exploitable patterns on Polygon. Our findings demonstrate how protocol-induced latency pressures can lead to catastrophic cryptographic failures in production blockchain systems, where a single implementation error compromises multiple accounts simultaneously.
arXiv:2605.22044v1 Announce Type: new
Abstract: Accurate localization of myocardial infarction is essential for risk stratification. While LGE-MRI remains the gold standard, it is resource-intensive. Integrating cine MRI with ECG enables a more detailed representation of infarct properties. Existing inverse MI inference methods overlook realistic scar morphology and cardiac repolarization, reducing sensitivity to subtle ECG variations and interpretability of infarct-induced electrophysiological changes. In this paper, we propose a novel framework for noninvasive MI localization using cardiac digital twins. To bridge the domain gap between simulation and reality, we introduce an anatomy-aware stochastic infarct synthesis strategy to synthesize realistic, irregular scars with border zones, mimicking ischemic transmural progression. We then construct a virtual cohort to simulate QRS-T waveforms, capturing both depolarization and repolarization dynamics. Furthermore, we design a Physiology and Anatomy Aware Network (PAA-Net) that jointly encodes 3D myocardial geometry and multi-lead ECGs to infer infarct areas with varying localizations, sizes, spatial extents, and transmuralities. Experimental results demonstrate that our framework significantly outperforms existing methods in inverse inference, achieving Dice scores of 0.7391 and 0.5503 for scar and border zone segmentation, respectively, while further enhancing the interpretability of the ECG-infarct relationship. Our code will be released upon acceptance.
arXiv:2605.21507v1 Announce Type: new
Abstract: Atmospheric visibility is a critical variable for transportation safety and air quality management, however, accurate prediction remains challenging due to the complex interactions between meteorological conditions and air pollutants, as well as the rarity of low-visibility events. This study introduces a machine learning framework to nowcast visibility in six major South Korean cities. To handle the imbalance in the 2018-2020 training data, we applied the Synthetic Minority Over-sampling Technique with Nominal and Continuous (SMOTENC) and Conditional Tabular Generative Adversarial Network (CTGAN). An ensemble approach combining machine learning and deep learning models was then used and evaluated on a 2021 test dataset. The results revealed a marked decline in predictive performance in the test set compared to the cross-validation phase. This degradation was attributed to a distributional shift between training and testing periods, which was quantitatively confirmed by measuring the Wasserstein distance of the most influential feature identified by SHAP analysis. In general, this study presents a methodology that aims to simultaneously address the dual challenges of data imbalance and temporal distributional shifts, and emphasizes the necessity of accounting for evolving external environmental factors when implementing nowcasting models on time-series data.
arXiv:2605.21509v1 Announce Type: new
Abstract: We use a newly commissioned rapid expansion aerosol chamber (REACh) facility to study the homogeneous nucleation of water vapor to form liquid droplets. We perform high-speed measurements to track the partitioning of water into vapor and droplets throughout the expansion process, including tunable diode laser absorption spectroscopy (TDLAS) to access the vapor concentration and in-line holography to track the size and concentration of nucleating droplets. We retrieve the peak saturation ratio achieved in each expansion from the TDLAS measurements in combination with adjusted thermocouple temperature readout. We monitor the number of nucleated droplets and their subsequent growth as a function of saturation ratio, and observe the onset of homogeneous nucleation of water vapor occurring at a threshold saturation ratio near $S=5$, in agreement with prior literature and classical nucleation theory. The trends we observe in average diameter and droplet concentration suggest that warm air pockets near the chamber walls inhomogeneously mix with cold air at the center of the chamber following expansion. Active forced mixing with fans yields more spatially uniform temperature readings across the chamber, but also significantly broadens the droplet size distribution. Our results demonstrate the capability of TDLAS and holography techniques to track both water vapor and liquid water in the high saturation ratio environments necessary for the homogeneous nucleation of droplets. Our findings also reveal that droplet nucleation and growth dynamics are highly sensitive to turbulence.
arXiv:2605.21822v1 Announce Type: new
Abstract: Reinforcement Learning from Human Feedback (RLHF) can reveal implicit objectives such as safety considerations that go beyond task completion. In this work, we focus on the common safety criteria embedded in crowd preference datasets, where different users may express distinct preferences or objectives, yet follow similar safety principles. Our aim is to discover shared safety criteria from crowd preferences and then transfer them to downstream RL tasks to regularize agent behavior and enforce safety. We first show that direct reward combination-optimizing a preference-learned reward model together with downstream task rewards-has inherent limitations. Motivated by this, we propose Safe Crowd Preference-based RL, a hierarchical framework that extracts safety-aligned skills from crowd preferences and composes them via a high-level policy to safely solve downstream tasks. Experiments across safe RL environments and a preliminary LLM-style task with diverse user goals and shared safety constraints demonstrate that our approach substantially lowers safety costs without access to explicit safety rewards, while achieving task performance comparable to oracle methods trained with ground-truth safety signals.
arXiv:2605.21914v1 Announce Type: new
Abstract: This paper presents a non-contact approach for vibration-based structural damage detection using an autonomous and customized cost-effective unmanned aerial vehicle (UAV). Vibration signals are extracted from video recordings through vision-based motion tracking to identify shifts in natural frequencies indicative of structural degradation. A laboratory-scale frame structure is evaluated under healthy and simulated-damage conditions. The proposed system is validated through an experimental study involving two smartphones, a USB camera, and a custom-built low-cost UAV equipped with an onboard camera and an autonomous alignment system for operation in GPS-denied environments. The displacement time is extracted and analyzed in the frequency domain and compared to reference measurements from contact accelerometers and a finite element model. Experimental results show that all platforms successfully capture the fundamental frequency and its shift due to damage. Although the UAV exhibits slightly higher errors (up to 5.7%) due to platform-induced disturbances and sensing limitations, it reliably detects damage-induced frequency changes. Compared to commercial UAV systems, the proposed platform achieves comparable inspection performance at significantly lower cost. These results demonstrate that low-cost autonomous UAVs provide a practical, flexible, and scalable solution for structural health monitoring, particularly in scenarios where contact-based sensing is impractical. The findings also support the potential for the deployment of multiple cooperative UAVs to further enhance inspection coverage and robustness.
arXiv:2605.21515v1 Announce Type: new
Abstract: LLM prompting is widely used for naturally stated tasks, yet it is unreliable it may succeed on a few test cases but fail at deployment time. We study performance prediction: given a program, either symbolic (e.g. Python) or a prompt executed on an LLM, and a few in-domain examples, predict its performance on unseen tasks from the same domain. We use a simple coin-flip model, treating each pass/fail program execution as a Bernoulli random variable, whose success probability is the programs unknown performance. In this model, performance depends entirely on: 1) the observed execution outcomes on test cases, and 2) a prior over performances. We compile empirical performance priors from a corpus of diverse programs and tasks, and find that performance for symbolic programs (e.g., Python) are all or nothing, while prompt programs have a diffuse prior with many nearly-correct programs. This difference explains why a few passing tests can certify symbolic programs but not prompt programs. Building on this insight, we develop RAP (Retrieved Approximate Prior), which retrieves similar tasks and prompt programs from an existing corpus to construct a proxy prior, which is then used to predict performance. We show RAP achieves solid performances.
arXiv:2605.21499v1 Announce Type: new
Abstract: Grid-based neural networks such as convolutional autoencoders are widely used in dimension reduction-based surrogate models for computational fluid dynamics. In recent years, the use of coordinate-based approaches like conditional neural fields has emerged. Their independence of the spatial discretization is a beneficial feature for various applications in computational fluid dynamics. This paper discusses the spatio-temporal prediction of aircraft ditching loads using a conditional neural field approach. The model is evaluated using two datasets for the dynamic loads of the fuselage of a DLR-D150 aircraft, one of which relates to a single fixed spatial discretization and the other that includes data from different discretizations. When paired with a long short-term memory (LSTM) network in the latent space, the neural field-based model achieves a spatio-temporal prediction accuracy for the first data set that is close to that of grid-dependent convolutional autoencoder-based models, and with significantly less parameters. Results for the second data set demonstrate the ability of the neural field-based approach to reconstruct ditching loads accurately for heterogeneous spatial discretizations. This allows for flexible use of training datasets generated for different geometries and/or discretizations, as well as the use of the surrogate model to predict loads for different configurations.
arXiv:2605.21503v1 Announce Type: new
Abstract: Along with the development of the scripting technology in cartography, such as the GMT and libraries of R programming language, geologic and geophysical mapping is being vigorously promoted, where the integration of the thematic data, such as GEBCO, EGM-2008 and geological raster and vector layers is one of the primary datasets that provides the high-resolution raw sources for cartographic visualization in the geologically complex regions like Pakistan. This study aims to integrate scripting methods of automated cartography, methods of applied geoinformatics for geomorphometric analysis and technical data processing (formatting, projecting, plotting), to provide a synthesis of the geological, geophysical and geomorphological maps of Pakistan they as new information supporting analysis of the geospatial variations of geology, geomorphology, tectonics and gravity fields with a special focus on the geologically remarkable region of Pakistan: Sulaiman Fold Belt and Muslim Bagh ophiolite complex. This study presents new 12 thematic maps, which are technically made using scripting approaches and open tools. All maps cover the region of Pakistan and they are made using open source tools: GMT, R and QGIS. A GMT and R based scripting mapping is applied for mapping Pakistan, and its algorithm steps are presented stepwise as code snippets. A system complex approach of the data integration and formats reshaping, data conversion and reformatting for a single project of the geology of Pakistan is designed and developed based on the combination of the programming and scripting techniques and with additional mapping, which integrates the geospatial datasets. Correlation between spatial phenomena of Earth's gravity, geologic evolution and tectonic movements were commented.
arXiv:2605.21512v1 Announce Type: new
Abstract: In this paper we discuss extensions of the canonical quantization procedure in quantum field theories. We focus specifically on S-matrix representation as a T-exponent. This extension involves flat bundles on certain infinite dimensional functional manifolds of local time. The motivating problem is first principles treatment of bound states in quantum chromodynamics as well as precision physics of hydrogen atom and the muonium. Our main results include systematic treatment of flat bundles in an infinite dimensional setting, generalization of Hamiltonian evolution and functional renormalization group evolution equations in quantum field theories. We discuss several results from finite dimensional theory that have analogies in the functional setting. This includes construction of moduli space of flat connections and isomonodromic deformations. One of the outcomes of our analysis is a construction of a rich family of functional flat bundles with rational connections. This class of connections exhibits a rich set of mathematical properties. In particular, we construct examples of spaces fundamental groups of which have a definable continuum of generators. Physical states correspond to points in the moduli space of bundles on these spaces. On the physics side of things, we conclude that spacetime notions, such as spaces of particle configurations, emerge effectively as spectral sets of functional differential operators.
arXiv:2605.21514v1 Announce Type: new
Abstract: Many complex systems can be modeled by temporal networks, whose organization often evolves through distinct structural phases. Detecting the change points that delimit these phases is both important and challenging. In this work, we extend the conditional entropy of heat diffusion from static graphs to temporal networks and study its properties. We provide an upper bound and explain how discrepancies from it arise from the presence of asymmetric temporal paths. Moreover, we show that this quantity is monotone in time, yielding an information-theoretic analog of the second law of thermodynamics for inhomogeneous diffusion on temporal networks. We then introduce a local version of conditional entropy, designed to probe diffusion over finite temporal windows, and show that it provides an informative signal for change-point detection in continuous-time temporal networks. We evaluate the proposed methodology on synthetic benchmarks, including comparative experiments with existing nonparametric baselines in the snapshot setting, and then apply it to a real-world temporal contact network from a French primary school. Finally, we show how to use detected change points to perform community detection on targeted sub-intervals, improving the quality and interpretability of the clustering results.
arXiv:2605.21552v1 Announce Type: new
Abstract: Confidence calibration for classification models is vital in safety-critical decision-making scenarios and has received extensive attention. General confidence calibration methods assume training and test data are independent and identically distributed, limiting their effectiveness under covariate shifts. Previous calibration methods under covariate shift struggle with class-wise or canonical calibrations and often rely on unstable importance weighting when density ratios are large or unbounded. Given the above limitations, this paper rethinks confidence calibration under covariate shifts. First, we derive a necessary and sufficient condition for confidence calibration under covariate shifts, named Expectation consistency condition, which reveals covariate shifts do not necessarily lead to uncalibrated confidence and provides a weaker condition for confidence calibration than global covariate distribution alignment. Then, utilizing Expectation consistency condition, this paper proposes an unsupervised domain adaptation loss to calibrate confidence of the target domain, named Expectation consistency loss (ECL), which is compatible with canonical calibration, class-wise calibration, and top-label calibration. Third, we prove that computing ECL loss has the same sample complexity as Expected Calibration Error (ECE) and provide a theoretically grounded mini-batch trainable scheme for ECL loss. Finally, we validate the effectiveness of our method on both simulated and real-world covariate shift datasets.
arXiv:2605.21566v1 Announce Type: new
Abstract: Machine learning models for chronic kidney disease (CKD) risk prediction often post strong discrimination scores on internal test sets. Calibration and uncertainty quantification get far less attention, leaving clinicians without reliable information about whether the probability outputs are accurate.
We trained five classifiers on the UCI CKD dataset (400 patients, 62.5% CKD prevalence): logistic regression, random forest, XGBoost, SVM with Platt scaling, and Gaussian naive Bayes. We evaluated each across calibration quality, conformal prediction coverage, and an eight-criterion deployment readiness framework. A distributional stress-test applied the best-calibrated variant of each model to the open-access MIMIC-IV demo cohort (97 patients, 23.7% CKD) to assess behaviour under prevalence shift and feature missingness. We measured calibration before and after Platt scaling and isotonic regression using Expected Calibration Error and Brier Score, and quantified uncertainty through split conformal prediction targeting 90% marginal coverage.
All five models reached AUROC 1.00 on the UCI test set. Isotonic recalibration reduced internal ECE to 0.000-0.022. On MIMIC-IV, AUROC fell to 0.48-0.58, ECE rose to 0.68-0.76, and conformal coverage dropped from 0.80-0.98 to 0.21-0.25 against a 90% target. No model scored above 4 out of 16 on the deployment readiness checklist.
Near-perfect internal performance did not transfer. Calibration stability and conformal coverage should be evaluated on external data before any clinical prediction model moves toward deployment.
arXiv:2605.21572v1 Announce Type: new
Abstract: Simulation-ready physical 3D assets have emerged as a promising direction owing to their broad applicability in downstream tasks. However, most existing 3D generation methods either neglect physical properties or are limited to a single asset category, e.g., rigid, deformable, or articulated objects. To address these limitations, we introduce PhysX-Omni, a unified framework for simulation-ready physical 3D generation across diverse asset types. Specifically, we develop a novel and efficient geometry representation tailored for Vision-Language Models, which directly encodes high-resolution 3D structures without compression, significantly improving generation performance. In addition, we construct the first general simulation-ready 3D dataset, PhysXVerse, covering diverse indoor and outdoor categories. Furthermore, to comprehensively and flexibly evaluate both generative and understanding capabilities in the wild, we propose PhysX-Bench, which encompasses six key attributes: geometry, absolute scale, material, affordance, kinematics, and function description. Extensive experiments with conventional metrics and PhysX-Bench show that PhysX-Omni performs strongly in both generation and understanding. Moreover, additional studies further validate the potential of PhysX-Omni for applications in simulation-ready scene generation and robotic policy learning. We believe PhysX-Omni can significantly advance a wide range of downstream applications, particularly in embodied AI and physics-based simulation.
arXiv:2603.03454v2 Announce Type: replace
Abstract: Offline Reinforcement Learning (RL) is an emerging field of RL in which policies are learned solely from demonstrations. Within offline RL, some environments involve balancing multiple objectives, but existing multi-objective offline RL algorithms do not provide an efficient way to find a fair compromise. FairDICE (see arXiv:2506.08062v2) seeks to fill this gap by adapting OptiDICE (an offline RL algorithm) to automatically learn weights for multiple objectives to e.g. incentivise fairness among objectives. As this would be a valuable contribution, this replication study examines the replicability of claims made regarding FairDICE. We find that many theoretical claims hold, but an error in the code reduces FairDICE to standard behaviour cloning in continuous environments, and many important hyperparameters were originally underspecified. After rectifying this, we show in experiments extending the original paper that FairDICE can scale to complex environments and high-dimensional rewards, though it can be reliant on (online) hyperparameter tuning. We conclude that FairDICE is a theoretically interesting method, but the experimental justification requires significant revision.
arXiv:2602.05304v2 Announce Type: replace
Abstract: Stochastic variance-reduced algorithms such as Stochastic Average Gradient (SAG) and SAGA, and their deterministic counterparts like the Incremental Aggregated Gradient (IAG) method, have been extensively studied in large-scale machine learning. Despite their popularity, existing analyses for these algorithms are disparate, relying on different proof techniques tailored to each method. Furthermore, the original proof of SAG is known to be notoriously involved, requiring computer-aided analysis. Focusing on finite-sum optimization with smooth and strongly convex objective functions, our main contribution is to develop a single unified convergence analysis that applies to all three algorithms: SAG, SAGA, and IAG. Our analysis features two key steps: (i) establishing a bound on delays due to stochastic sub-sampling using simple concentration tools, and (ii) carefully designing a novel Lyapunov function that accounts for such delays. The resulting proof is short and modular, providing the first high-probability bounds for SAG and SAGA that can be seamlessly extended to non-convex objectives and Markov sampling. As an immediate byproduct of our new analysis technique, we obtain the best known rates for the IAG algorithm, significantly improving upon prior bounds.
arXiv:2605.21622v1 Announce Type: new
Abstract: Topology optimization can generate efficient structures, but designers often must manually translate qualitative intent, such as desired visual style, product experience, or manufacturability into solver settings that are not directly tied to those preferences. We present TO-Agents, a multi-agent AI framework that connects natural-language design intent with iterative topology optimization. The framework converts a human-provided problem description into validated solver inputs, runs a topology optimization solver, renders the resulting 3D topology, and uses multi-view vision-language reasoning with an independent judge agent to critique each result and revise solver parameters. We evaluate the framework on two long-horizon design tasks: a cantilever beam benchmark and a phone-stand product design. In both tasks, the designer specifies an aesthetic preference for hierarchically branched structures inspired by natural tree morphologies, and the system performs four revision cycles across ten independent replicates. TO-Agents produces at least one preference-aligned design in 60% of trials for each case study, corresponding to up to 6x more successful trials than an ablated pipeline without visual or historical feedback. Judge scores and human evaluations show that the pipeline can identify effective parameter levers, recover from poor revisions, and expand design exploration. A manufacturing agent further post-processes top-ranked designs for additive manufacturing, enabling end-to-end intent-to-prototype design. We also identify failure modes, including overshooting, selective memory, misplaced tools, and incorrect parameter reasoning. These results suggest that agentic topology optimization can shift designers from low-level parameter tuning toward higher-level specification of form and function, while highlighting safeguards needed for reliable autonomous engineering design.
Specification-Driven Generation and Evaluation of Discrete-Event World Models via the DEVS Formalism
arXiv:2603.03784v2 Announce Type: replace
Abstract: World models are central to LLM agents that must evaluate actions over long horizons. Yet much existing work focuses on environments governed by physical dynamics or spatial structure, whereas many high-impact domains, including supply chains, procurement networks, and business processes, evolve through discrete events, timing constraints, and causal dependencies. These settings call for discrete-event world models. Existing approaches to constructing world models often fall near two extremes: hand-engineered simulators provide consistency and reproducibility, but are costly to build and adapt; neural models are flexible, but can suffer from compounding inconsistency over long-horizon rollouts. We seek a principled middle ground by synthesizing discrete-event world models online from natural-language specifications, retaining the reliability of explicit simulators while gaining the adaptability of neural models. We adopt the DEVS formalism and introduce a staged LLM-based generation pipeline that separates structural inference over component interactions from component-level event and timing logic. For evaluation, we develop benchmark suites in which simulators emit structured event traces, which are then validated against specification-derived temporal, causal, and semantic constraints. This enables reproducible verification and localized diagnostics. Together, these contributions produce world models that remain consistent over long-horizon rollouts, can be verified from observable behavior, and can be synthesized efficiently on demand during online execution.
arXiv:2605.22040v1 Announce Type: new
Abstract: Fluid antenna systems (FASs) are emerging as a reconfigurable-aperture technology that expands physical-layer design beyond fixed, rigid antenna geometries. While the \emph{fading diversity} of FASs -- which exploits spatial channel fluctuations for signal enhancement and interference avoidance -- has been widely studied, the \emph{geometry diversity} created by reconfigurable port placement remains far less understood, particularly for planar architectures under finite-aperture constraints. This paper develops a systematic analytical framework for finite-aperture planar fluid antenna arrays (FAAs). First, we derive a closed-form characterization of the minimum inter-port distance under uniform random placement over a rectangular aperture and show that it follows a Rayleigh law. Its mean scales as $\mathcal{O}(M^{-1})$, in sharp contrast to the $\mathcal{O}(M^{-2})$ behavior in the linear case in which $M$ represents the number of candidate ports, revealing a fundamentally more favorable packing geometry in two dimensions. Secondly, we establish a universal Cram\'{e}r-Rao bound (CRB) for joint elevation-azimuth estimation, governed by a $2\times 2$ \emph{geometric inertia matrix} whose determinant and eigenstructure fully capture the role of port placement in estimation precision. We further prove that both the trace and determinant of this matrix are invariant to the azimuth look direction. Third, we uncover an intrinsic \emph{precision--ambiguity trade-off}: maximizing the geometric determinant to minimize the CRB drives ports toward the aperture boundary, but simultaneously increases sidelobe-induced spatial ambiguity.
arXiv:2605.21661v1 Announce Type: new
Abstract: Adapting pretrained diffusion models to downstream objectives such as inverse problems often requires expensive test-time guidance or optimization. We propose a principled framework for generating high-quality reward-aligned samples at substantially reduced inference cost. Our approach formulates test-time adaptation as a hierarchical variational model, where control is amortized into a lightweight yet expressive stochastic policy. This formulation naturally supports few-step diffusion sampling: large step sizes enable fast inference, while the learned policy maintains sample quality by providing structured per-step control. The resulting fully amortized sampler achieves a strong quality--speed tradeoff, matching or exceeding recent test-time scaling baselines while requiring significantly less compute. For example, on 4x super-resolution, our method achieves better perceptual quality with more than 5x faster inference compared to the best-performing baseline. We further extend our approach to a semi-amortized regime that combines cheap amortized proposals with limited test-time optimization, achieving state-of-the-art perceptual quality across several challenging inverse problems.