Papers

The theoretical description of nuclear structure in rare earth nuclei represents a significant challenge for many shell models. This difficulty stems from the size of the configuration space involved. Methods based on symmetries provide significant advantages. Furthermore, the pseudo-SU(3) shell model has proven to be very useful in the description of these systems. In particular, we propose to study the energy spectra of the rotational bands with normal deformation at low and medium spin ($J\leq35/2$), the quadrupole moments, and the deformation parameter $\gamma$ associated with the states, as well as the B(E2) transition strengths in the $^{167,169,171}$Lu nuclei. The Hamiltonian includes $Q \cdot Q$, Nilsson, and pairing terms, parameterized in a systematic way, in addition to three rotor-like terms that enable fine tuning of the spectra. Our calculations predict collective rotational bands with prolate normal deformation in most cases. The exception is the $1/2^+$ band in $^{171}$Lu with a triaxial shape. The quadrupole moments vary along the bands, increasing in absolute value with spin. The $\gamma$ deformation parameter remains nearly constant along the bands, and the B(E2) values suggest bands with collective character and normal deformation. The agreement and limitations of the model are discussed.

Terrain assessment is a critical capability for off-road mobile robots, enabling safe and reliable navigation through unstructured and geometrically complex environments. Conventional geometry-based terrain assessment is fast to compute but often overly conservative in unstructured environments. We present PIVOT: a Physically Informed Vision-Language Off-Road Traversability navigation system that augments conventional geometry-based planning with vision-language-model (VLM)-based semantic reasoning for field robots. To physically ground this assessment, we quantify how strongly the VLM's predicted traversal energy cost, robot vibration, and wheel slip correlate with real-world measurements and introduce a unified traversability score that weights each modality by its prediction-measurement correlation. For efficiency, we design a two-level navigation architecture that retains geometry-based planning as the nominal mode and invokes semantic replanning only when that mode fails to find a path. Across five repeated closed-loop trials on a mixed-terrain route totalling around $6.4$ km, the proposed system increases overall autonomy from $59.6\%$ to $97.0\%$, reduces human interventions from $11$ to $3$, and increases the mean distance between interventions (MDBI) from $69.2$ m to $412.9$ m compared with geometry-only navigation. These results demonstrate that physically grounded VLM-based terrain assessment can substantially extend autonomous navigation beyond the limitations of geometry alone, while preserving efficient geometric planning as the nominal mode.

Reinforcement learning (RL) controllers have been recently adopted for Unmanned Aerial Vehicles (UAV) navigation and control. However, they are susceptible to action-space attacks that overwrite the action commands after the policy generates them and before the actuators execute them. While most existing defenses target attacks on the policy's inputs, those addressing action-space attacks retrain the policy at training time and are not resilient to corrupted actions at runtime. We propose ASGARD, a two-phase teacher-student pipeline for making RL-based UAV control resilient to action-space attacks. In the teacher phase, an encoder combines the UAV's physical state with action-attack-related privileged information to produce an action-attack-aware latent that trains the RL control policy and a monitor that outputs corrected action commands to the actuators. In the student phase, both the encoder and the monitor are trained via supervised learning from their teacher counterparts to run on-board using only the UAV's physical state history. We evaluate ASGARD across attack scenarios targeting different action commands on UAV. We find that ASGARD is resilient to action-space attacks and completes the missions despite the attack. We further find that ASGARD generalizes to unseen attacks and remains resilient against stealthy attacks.

Autoregressive (AR) models suffer from local greediness, while diffusion language models (DLMs) often lack the strict causal structure required for reasoning. To combine the advantages and overcome the drawbacks of the dual, we propose Causal Latent Revision (CaLR), a framework that reformulates reasoning as constrained latent optimization. By adopting a causal topology matrix (CTM) from an expert model and implicit differentiation, CaLR performs gradient-guided “thought revision" to enforce logical consistency, enabling dynamic self-correction of intermediate steps during parallel generation. Empirically, CaLR achieves SOTA DLM performance on complex benchmarks, surpassing strong AR baselines and demonstrating superior robustness in constrained tasks like Sudoku.

Vision-language-action (VLA) models have demonstrated strong capabilities in robotic manipulation, yet their reliance on visual perception limits robustness in contact-rich environments, where critical physical interaction states may not be visually observable. Existing tactile-enhanced VLA methods improve physical grounding using observed tactile feedback, but most remain largely reactive rather than explicitly modeling how contact may evolve. Therefore, we propose ForeTac-VLA, a forecasting-based tactile-vision-language fusion model that predicts future tactile states to guide action generation. Specifically, ForeTac-VLA encodes recent tactile observations into temporal representations and integrates them with vision-language features through bidirectional cross-attention. Further, a transformer-based forecasting module predicts multi-step future tactile states, enabling the model to reason jointly over observed and anticipated contact. Finally, the fused multimodal representations and predicted future tactile states are fed into the VLA backbone to condition action generation. To stabilize training, a ground-truth-to-prediction curriculum is employed when early forecasts are unreliable. Across four real-world contact-rich manipulation tasks, ForeTac-VLA achieves an average success rate of 95%, outperforming the fine-tuned VLA model by 36.25 percentage points and state-of-the-art tactile-enhanced VLA baselines by over 22 percentage points. ForeTac-VLA also maintains strong performance under low-illumination and visually cluttered conditions. Video demonstrations can be found on https://foretac-vla.github.io/

We give an elementary proof of the Komlós conjecture by simplifying the recent proof of Guo, Fang, and Lu. We show that any vectors $v_1,\ldots,v_n\in\mathbb{R}^d$ with $\|v_i\|_2\le1$ admit signs $\varepsilon_i\in\{-1,1\}$ such that $\|\sum_{i=1}^n\varepsilon_i v_i\|_\infty\le36$. The proof uses only elementary combinatorial and probabilistic arguments and basic calculus.

High-consequence subsurface decisions often rely on sparse data that permit competing geological interpretations. Determining consistency of these interpretations with the available observations remains challenging. We present a workflow that addresses this challenge by translating competing geologic interpretations into alternative priors and ranking them according to their consistency with hydraulic-head observations. A key step in this workflow is exploiting the broad knowledge of image-generation foundation models to transform nuanced geologic interpretations into data ready for computer modeling. For each interpretation, a text-to-image foundation model generates an ensemble of geologic images, and a separately trained variational autoencoder learns an interpretation-specific latent representation. A supervised inverse network maps head observations into this latent space, and the frozen decoder reconstructs an image that is mapped to a log-conductivity field. Steady-state flow simulations predict heads, and the aggregate normalized head error determines the ranking. We evaluate the framework using a synthetic benchmark based on the Johansen Formation with three interpretations of decreasing consistency with the reference geology. Across 595 test cases, the Precise & Accurate interpretation produces lower normalized errors than Accurate in 58.5% of cases and Mismatched in 82.5% of cases. Accurate outperforms Mismatched in 65.5% of cases. We then compare spatial representations of two published conceptual models of the Culebra Dolomite Member at the Waste Isolation Pilot Plant. The revised representation yields an aggregate normalized error of 7.598, compared with 8.595 for the original, consistent with the documented conceptual-model revision. The framework enables quantitative comparison of competing geological interpretations using available hydraulic observations.

AI-enabled software development tools (AI-devtools) are being industrially adopted under strong expectations of productivity gains, yet developers' experiences of such roll-outs are underexplored. Organizations commit budgets, evaluate staff, and revise practice on a partial picture, since the evidence base is mainly tool evaluations, productivity metrics, and surveys, with few qualitative in-situ accounts of ongoing, coordinated roll-outs. We report a case study of a coordinated roll-out of AI-devtools at a large Swedish telecommunications company, investigating how developers experience the roll-out and how they anticipate their profession will change. We conducted semi-structured interviews with 12 software professionals across three sites, analyzed with process coding and thematic analysis, and interpreted through the extended Technology Acceptance Model (TAM2) as a post-hoc analytical lens. Our findings on use cases, productivity, frustrations, and tool limitations corroborate prior survey work. Beyond corroboration, the interviews surface a management-developer expectation gap that maps onto the TAM2 constructs of subjective norm and voluntariness, and show that participants weigh perceived risk heavily, a factor that TAM2 and similar acceptance models do not represent. AI-devtools emerge as helpful but fallible assistants whose value is shaped by organizational expectations, system scale, and developers' skills.

The amplitude and waveform shape of gravitational wave radiation from compact binary mergers enable a direct measurement of the luminosity distance to these events, making them “standard sirens”. Binary black hole (BBH) mergers form a significant fraction of these standard sirens. A substantial catalog of BBH merger events can be used to probe cosmology and large-scale structure properties, such as the baryon acoustic oscillations (BAO). In this work, we explore detecting the BAO scale using the cross-correlation between BBH mergers and a galaxy catalog, and compare it with the BBH autocorrelation for the third-generation detector observations. Over a $10$-year observation period, these observatories will localize approximately $1.5\times10^{5}$ events within $10$ square degrees, with \snr $\geq 50$ and luminosity distance uncertainty $\leq 50$ Mpc, and will allow recovery of the BAO scale. We find that the BAO scale is recovered with approximately two times the \snr and a significantly better precision through cross-correlation as compared to autocorrelation.

Affordable dust monitoring remains a pressing need for the cattle feedlot industry, yet camera-based PM estimation, despite its growing body of research in urban air quality settings, has not been evaluated under the extended concentration ranges characteristic of intensive livestock operations. This study developed an image-based approach using contrast panel features and machine learning to estimate PM10 concentrations in a commercial cattle feedlot, where hourly average PM10 ranged from 250 to 1,000 ug/m^-3 and instantaneous concentrations reached 5,000 to 20,000 ug/m^-3. Grayscale images were captured during the evening dust peak period, and features including panel contrast, black and white panel pixel values, and overall image brightness were extracted. The model also incorporated recent past values from preceding images and solar zenith angle as predictors. Among the candidate models evaluated, XGBoost achieved the highest predictive performance, with an R^2 of 0.792 and a median absolute error of 103 ug/m^-3. Feature importance analysis revealed that (a) panels positioned farthest from the camera contributed most strongly to predictions and (b) that black panel pixel values were more sensitive than white panel values to changes in PM10 concentration. Prediction accuracy during the sunset transition, which coincides with the onset of the feedlot evening dust peak, remains an area for further refinement. These findings demonstrate the feasibility of image-based PM10 estimation across PM concentration ranges substantially exceeding those reported in prior urban studies and provide practical guidelines for future deployment in feedlot environments.

In Mixture-of-Experts language models, the router typically selects and weights experts based on the token's hidden state, utilizing limited contextual information. We propose Attention-Aware Routing (AAR), which augments the router with temporal and spectral features extracted from a sliding window of attention weights that represent a summary of the model's contextual state, disentangled from the hidden state. Keeping the base transformer entirely frozen, we train only the routing parameters, isolating routing as the sole variable. AAR improves GSM8K by +3.37 pp over a routing-only SFT baseline on OLMoE. Beyond performance, we show that routing and attention form a coupled circuit: routing changes at layer l propagate through the residual stream to amplify attention sinks at layer l+1, reshaping attention without any direct update to the attention mechanism itself. Further, AAR reduces long diverging generation, with incorrect answers getting shorter, while correct answers remain unchanged in length. Finally, AAR is strongly depth-sensitive: applying it indiscriminately across layers can degrade factual retrieval, whereas mathematical reasoning gains persist when it is introduced deeper in the network. This sensitivity exposes a retrieval–reasoning tension across depth and makes layer-selective AAR a controlled probe of the routing-relevant information carried by attention at different layers.

Complex problem solving (CPS) with large language models (LLMs) is often framed as a matter of stronger reasoning or longer generation. Yet early-step error amplification, prompt brittleness, and failures to revise incorrect commitments are difficult to explain by missing knowledge or expressive capacity alone. This survey interprets CPS as a sequential estimation-and-decision problem over a latent solution state. A controller maintains a belief about an unobserved solution trajectory, updates it as noisy intermediate evidence arrives, and decides whether to commit, verify, branch, roll back, or abstain to minimize expected loss. Reasoning supplies candidate transitions and interpretations, whereas process control shapes and evaluates those proposals and regulates subsequent transitions and observations. Within this framework, we organize existing methods around five components: explicit state representation, transition structuring, validation and constraint enforcement, search and rollback, and uncertainty management. We also interpret evaluation metrics according to the statistical quantities they estimate. The framework further yields a diagnostic hypothesis: interventions should be most effective when they target the error or uncertainty component implicated by an observed failure. We distinguish systematic, stochastic, and irreducible error together with epistemic and aleatoric uncertainty, and call this alignment problem-control fit and its failure control mismatch. For example, additional sampling may reduce sampling variability while leaving a shared systematic error unchanged. This perspective clarifies what current methods estimate and control, what remains uncontrolled, and why reliable validation, targeted recovery, calibrated uncertainty, and matched-budget evaluation are central open problems.

State-of-the-art key-value stores built for persistent memory (PM) provide low latency as they allow application threads to directly access the data on PM and rely on hardware to manage multiple non-volatile DIMMs (NVDIMMs). While this provides low latency, it results in low throughput and scalability. Performance degrades because PM hardware requires fine-grained control over PM accesses; for example, throughput degrades if too many threads write to PM concurrently. We present SKYE, a write-optimized PM key-value store that achieves high throughput and scalability. SKYE builds on the central idea of maintaining fine-grained control over all PM accesses and obtains high PM write-bandwidth utilization. To achieve this, SKYE deviates from current practice and provides indirect access to applications; applications send requests to SKYE, which uses dedicated threads to access PM on their behalf. Instead of relying on hardware-managed PM, SKYE controls how data is placed on individual NVDIMMs. SKYE leverages multiple media to avoid overloading PM and limits remote NUMA accesses for scalable throughput. We show that on a single NVDIMM, SKYE outperforms state-of-the-art PM stores by 2.5-5x on the standard Yahoo Cloud Serving Benchmark (YCSB). With four NVDIMMs across four NUMA nodes, SKYE obtains about 86% of PM write bandwidth, and its write throughput scales by 3.9x.

Long-context large language model inference is increasingly limited by prefill, where dense self-attention processes the entire prompt before generation begins. Sparse block selection can reduce this cost, but a block centroid may hide a highly relevant token among many irrelevant ones. We call this failure mode mean dilution and propose RBS-Attention, a training-free sparse-prefill method with two complementary selection branches. A centroid base branch captures average relevance, while a rescue branch uses the maximum key-block radius and its prompt-, layer-, and head-dependent distribution to identify blocks at risk of underestimation. Independently thresholding the two branches and combining their masks controls the contribution of rescue blocks while preserving regular block-sparse FlashAttention execution. On H100 GPUs, RBS-Attention achieves 20.65$\times$ standalone prefill-attention speedup, 11.92$\times$ vLLM prefill-attention speedup, and 5.97$\times$ end-to-end time-to-first-token speedup at 128K on Qwen3-30B-A3B-Instruct-2507-FP8. On the dense Qwen3-32B model, it obtains 88.65 overall RULER accuracy versus 89.52 for dense attention; LongBench-v2, InfiniteBench, and Video-MME provide additional quality evaluation. Supporting experiments measure actual retention, compare selectors at matched density, and characterize block-size, threshold, and memory behavior. Together, these results support radius-adaptive dual-branch selection as an effective approach to long-context prefill.

Many physical properties relevant to robotic manipulation are hidden from vision. A sealed object, for example, may reveal little about its center of mass (COM) or internal contents until it is lifted, shaken, or otherwise dynamically perturbed. This study shows that such interactions can enable a new modality of robotic perception and learning, in which interaction-induced dynamic responses are used to infer object physics that is inaccessible to conventional sensing. We implement this idea using an origami-inspired soft robotic arm that functions as a physical reservoir computer. After grasping an object, the arm is excited by a fixed shaking input at its base, and the resulting ringdown response is recorded through either camera tracking or embedded sensors. Because the input is held constant across trials, hidden object properties, such as the COM position, are encoded through their effect on the dynamics of the coupled robot-object system. A lightweight linear readout can then decode these dynamics to recover interpretable information about the hidden object physics. Using this framework, the soft robotic arm reservoir completed three tasks of increasing difficulty: inferring the orientation of the object's hidden COM, inferring the COM distance from the grasp point, and using the inferred COM information to guide a subsequent regrasp. We further develop a dynamic summary representation of the ringdown response that improves prediction accuracy. Together, these results establish shake-to-learn mechanical interrogation as a promising strategy for robotic systems to convert brief physical interactions into actionable cues about hidden object properties for downstream manipulation.

The aim of this work is to develop a self-contained existence and uniqueness theory for a doubly nonlinear parabolic problem with variable exponents and homogeneous Neumann boundary conditions. Our approach is based on Rothe's time-discretization method and builds on the results established by the author in a previous paper. We also investigate the asymptotic behavior of the solution. All arguments are presented in full detail, so that the paper is self-contained.

In non-stationary online learning, dynamic regret has attracted increasing attention as a measure of how well an online learner performs against a time-varying comparator sequence. Despite considerable advances, attaining optimal bounds for strongly convex and exp-concave losses often involves intricate analysis. In this paper, we present a simple framework that reduces dynamic regret minimization to switching regret minimization. As a result, we can derive dynamic regret bounds by using off-the-shelf algorithms with switching regret guarantees. The key idea of our reduction is to construct, for any comparator sequence, an auxiliary random sequence that is unbiased at each round, with the controlled variance and a manageable number of switches. Combining this construction with suitable surrogate losses, we can decompose dynamic regret into the expected switching regret against the random sequence and its controlled variance. Theoretically, for strongly convex and exp-concave losses, we establish the $\widetilde{O}(T^{1/3}P_T^{2/3})$ dynamic regret bounds, where $T$ denotes the time horizon and $P_T$ denotes the path-length of the comparator sequence. Moreover, for general convex losses, the same reduction also recovers the $O(\sqrt{T(1+P_T)})$ dynamic regret bound. Notably, all our findings match the minimax optimal results for these three types of losses, highlighting the versatility of our proposed framework.

We study frieze patterns over the integers that are allowed to have wild entries. We introduce the quiddity number as a new invariant. The quiddity number is then used to classify strongly connected components of the directed graph $\Gamma_{2,n}(\mathbb{Z})$. Furthermore, we show that every finite simple directed graph arises as an induced subgraph of a directed graph $\Gamma_{2,n}(\mathbb{Z})$ for $n$ sufficiently large.

Realistic work extraction from many-body quantum batteries must be local. However, only a small fraction of energy eigenstates of a generic many-body system, called scars, can support local extraction. The remaining bulk is useless for the task due to the eigenstate thermalization hypothesis. Here we devise a universal low-complexity protocol that steers any initial state towards exactly one scar—representing a charged state of the battery—from which a macroscopic amount of work can be extracted using local unitary operations. This is achieved by leveraging the nontrivial interplay of engineered dissipation and continuous indirect measurement that, in addition, leads to enhanced stability and charging speed compared to any other strategy using these processes independently. Moreover, the protocol works directly on the hardware level, in that it requires no simulation or suppression of interactions between subsystems. Our construction thereby enables macroscopic charge storage in steady states of generic nonintegrable many-body systems indefinitely, from which a reliable stream of work can be extracted via purely local means.

Close-proximity pipeline inspection is challenging for standard drones due to high hovering power consumption, airflow sensitivity, and propeller wash interference with gas sensing. To address this, we present AeRove, a compact bimodal aerial-terrestrial robot. AeRove uses its propeller guards as wheels to roll along pipes and employs a spring-loaded bistable mechanism to reconfigure between ground and flight modes in 200 ms without continuous actuator power to maintain either state. The converging propeller-guard geometry also improves measured thrust efficiency. By reserving flight for obstacle hopping and using ground rolling for continuous traversal, current draw is reduced 14x compared to continuous flight (0.7 A vs. 10 A), extending estimated travel distance from 144 m to 2057 m. Autonomous trials on a 51 cm diameter steel pipe demonstrated navigation on straight and curved sections, obstacle jumping, and leak detection of simulated inspection markers. Separate CO2 sensing experiments evaluated gas-detection performance under perched and aerial operating conditions. Perched inspection produced a substantially larger concentration response than hovering under the tested conditions. During flight, an underbody propeller-intake configuration produced a faster and stronger gas-detection response than an extended probe by leveraging propeller-induced airflow. Code and designs are released under the CC-BY license.

Modern inverse problems and adaptive simulations require efficient and differentiable representations of complex geometries that can be updated frequently for gradient-based optimization. Existing methods cannot simultaneously achieve both. We present Warp-Geo, a differentiable GPU-accelerated framework for reconstructing signed distance fields and evaluating surface normals from point clouds of complex 3D geometries. Warp-Geo combines uniform grids for GPU parallelism with implicit differentiation through the Poisson solve, enabling frequent geometry recomputation while maintaining full differentiability. We demonstrate Warp-Geo on forward SDF reconstruction, dynamic fluid–structure interaction with moving boundaries, and inverse shape optimization via automatic differentiation with gradient validation. GPU scalability analysis confirms practical feasibility for frequent geometry updates. The framework, implemented in Warp and JAX, enables seamless integration with differentiable physics solvers for end-to-end shape optimization.

Population III (Pop III) stars, the first generation of metal-free stars, remain undetected, partly because entirely pristine galaxies represent a short-lived phase. We instead target the longer-lived self-polluted and hybrid Pop III-Pop II phases, in which surviving metal-free Pop III stars coexist with enriched gas and newly formed Pop II stars, producing strong He II alongside detectable UV metal lines. We search for this signature in 84 strong He II-emitting galaxies at z = 3-7 in the JWST/JADES GOODS-South and GOODS-North fields, combining UV emission-line diagnostics, gas-phase metallicities, AGN screening, and spatially resolved SED fitting. We find no evidence for any pristine Pop III galaxies. Instead, The Blueberry (ID 13176, z = 5.94) is our strongest candidate for a hybrid Pop III-Pop II system and is the only source with robust detections of all diagnostic UV lines. Its resolved stellar populations are extremely young, blue, and weakly attenuated, with high star formation rates, consistent with intense recent star formation. While these properties alone cannot distinguish it from a chemically enriched starburst, its robust UV metal-line ratios provide the strongest evidence in our sample for a surviving Pop III contribution within an enriched host. The Carrot (ID 13577, z = 5.57) further demonstrates the power of resolved analysis, revealing a central region substantially bluer and younger than the galaxy as a whole. Together, these results suggest that Pop III signatures may be spatially localised and diluted in integrated measurements, highlighting resolved SED fitting as a tool for identifying environments where the Universe's first stars may survive.

The proliferation of harmful internet memes poses a significant societal threat, yet their automated classification remains a formidable algorithmic challenge due to the nuanced, multimodal nature of their content. To address this, we introduce MemeTAG, a novel dual-objective framework that pioneers a keyword-aware approach to meme classification. Our core innovation is a two-part semantic guidance mechanism: first, we leverage a pretrained Vision-Language Model to generate a set of descriptive keywords, that capture the high-level semantics. Second, we introduce the Aggregated Tag Inference Network (ATIN), an attention-based module that distills these keywords into a single, rich semantic embedding. This embedding serves as a target for a novel auxiliary reconstruction loss, which compels the model to learn deeply aligned visual and textual features. This approach, combined with an efficient three-stage training strategy, establishes a new state-of-the-art on the HarMeme, Hateful Memes Challenge (HMC), and PrideMM datasets, decisively outperforming existing state-of-the-art methods.

We prove that the independence sequence of every sufficiently large forest is unimodal. The result follows from establishing log-concavity on a central interval of the sequence, along with monotonicity of the initial and final segments. The main analytic step is a central limit theorem for the size of a random independent set sampled with the hard-core model, uniform over all forests and over an interval of positive fugacities.

We use a point-particle effective field theory to study the superradiant instability rates of an ultralight massive vector around a slowly rotating black hole. Earlier effective-theory treatments reported an apparent $O(1)$ discrepancy with the rates obtained by solving the Proca equation on a fixed black-hole background, e.g. for the mode with $j=1$ and $\ell=2$. We show that there is in fact no discrepancy: once the leading-order states in the degenerate $\{\ell=0,\ell=2\}$ subspace are correctly identified, the effective theory (implemented here with the graviton integrated out exactly, and expanded in $\alpha$ only afterwards) reproduces all of the $j=1$ Proca instability rates exactly, and it does so without ever invoking the FKKS ansatz on which the full-theory calculation rests. Since it never relies on separability, the same approach applies, at least in principle, to rotating backgrounds and bodies that lack the hidden symmetries underlying that ansatz. The single ingredient beyond the far zone is a conservative worldline contact term, the static monopole response of the horizon in the parity-even $j=1$ Proca sector, which we fix by matching to the near zone. The same term shifts the fine-structure splitting of the bound states into exact agreement with the published Proca spectrum. In addition to providing a conceptually straightforward model of vector superradiance and resolving an apparent tension in the literature, we present this work to illustrate potential technical subtleties involving dissipative effects in point particle effective theories.

The EU AI Act mandates that datasets for high-risk machine learning (ML) systems meet strict quality criteria such as soundness and bias mitigation. While Satisfiability Modulo Theory (SMT) solving offers a formal approach to verifying these properties, its scalability in realistic ML settings remains unexplored. To bridge this gap, this work presents the first large-scale empirical study of SMT-based dataset verification on two real-world ML datasets. We systematically evaluate how solver performance is shaped by three key dimensions: the type of data-quality property, the specification style, and the dataset encoding strategy. Our findings demonstrate that SMT-based verification is feasible for practical scenarios, but each dimension shapes it: the property type sets the tractability limit, the specification style drives scalability (exceeding $2{,}000{\times}$ for aggregate properties), and the encoding strategy has a systematic effect, with extracted feature columns performing best.

The coupling between layered atomically-thin materials mediated by van der Waals forces allows strong electronic correlations, unique topological signatures, and non-trivial spin textures, all of which play an important role in the creation of spin-, valley-, and orbitronic devices. Here, we demonstrate that twisted bilayer graphene encapsulated by transition metal dichalcogenides exhibits a generically non-radial spin texture yet hosts a purely collinear Edelstein effect despite this lack of radial symmetry. Moreover, we show how uniaxial pressure can be used to further tune the band structure and change the number of active Fermi surfaces without compromising the collinear response. Lastly, we illustrate how the quantum geometry changes in the encapsulated twisted graphene bilayer with larger pressures spreading the Berry curvature over large regions of the moiré Brillouin zone. These results illustrate how encapsulation and pressure can be used to drastically alter the topology and spin-charge interconversion processes of moiré heterostructures.

Performance models of LLM servers support both latency evaluation and the design of controllers for autoscaling against service level objectives (SLOs) and for inference optimization. We model the multiplexed execution of prefill and decode operations with a tractable, approximate queueing model under Markovian assumptions. Three parameters characterize a model-accelerator pair, namely a baseline per-iteration overhead, a per-token compute cost, and a per-token key-value (KV) cache access cost. The model combines a mean-value analysis of per-iteration work with a state-dependent Markov chain for batch occupancy to predict mean time to first token (TTFT) and inter-token latency (ITL). We validate these predictions against measurements and show that the three parameters can be estimated from observed latencies. Over a grid of input and output lengths and arrival rates spanning light to moderate load, the relative error of the average ITL is about 5% for Llama-3.1-8B and 8% for Qwen2.5-14B running on an H100 GPU, and the corresponding TTFT errors are 14% and 16%. We then implement an autoscaling controller that uses the model to adjust inference-server replica counts as the workload changes. On an OpenShift cluster of H100 GPUs it tracks a fourfold load ramp under both latency targets, missing one in 7 of 127 control cycles, and its in-loop predictions carry median errors of at most 5% for TTFT and 9% for ITL. A decode-throughput analyzer from an existing autoscaler, which takes no latency target, misses 27 of 128 cycles under the same controller and load while provisioning 4% and 28% fewer replicas.

Control of topological charges of bound states in the continuum (BICs) is essential for advanced topological photonics. A rigorous theoretical framework for understanding the formation of these charges is therefore necessary for further progress in this field. However, conventional multipolar formalism often fail to predict the topological charge when no single multipole dominates the mode. To address this fundamental issue, we present a rigorous dynamical systems framework for describing formation of topological charges. This attitude provides a direct link between polarization vortex and the local structure of the vector polarization field. Within the proposed framework, we challenge the established identification of the topological charge with the Hopf index of the dominant multipole and demonstrate the significant role played by derivatives of the multipolar coefficients in determining the charge. Our theoretical framework is validated within semi-analytical multipolar decompositions and full-wave numerical simulations of periodic dielectric metasurfaces. Furthermore, we show how the symmetry of the unit cell influences the local structure of the polarization field around BIC and identify the necessary conditions for the formation of high topological charges. Our findings provide a more rigorous basis for analysing BICs' topological properties, paving the way for advanced applications in topological photonics.

Projection-based controllers can overcome fundamental limitations of classical linear time-invariant control by modifying the controller's input-output behavior via projection. A key example is given by the hybrid integrator-gain system, a projected integrator, which has recently found successful application in several industrial systems. While prior work on analysis and design of projection-based control systems has primarily focused on the continuous-time setting and non-incremental analysis, a more refined incremental analysis in discrete-time is needed to better reflect actual digital implementation and obtain more accurate (robust) performance assessment. To address this need, this paper considers incremental stability and convergence analysis of discrete-time projection-based control systems. Our first methodology is based on showing that such controllers preserve the quadratic incremental stability of their nominal (unprojected) dynamics, if the projection metric is well-designed. Building on this, we derive a small-gain condition guaranteeing incremental input-to-state stability for interconnections of projected controllers with general nonlinear plants. A second approach is grounded in a direct Lyapunov-based method for verifying incremental stability in input-affine piecewise-smooth systems, which can be seen as an extension of the classical discrete-time Demidovic conditions. We illustrate our results through several examples, and demonstrate performance quantification via nonlinear Bode plots, with a special focus on first-order projection elements.