High Energy Physics (Experiment)
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 9 November
We introduce AlphaHits, a Transformer-based approach towards hit-level foundation models for LHC physics, building on the TrackFormer and HiggsFormer architectures. AlphaHits operates directly on detector hits and demonstrates scalable full-event charged-particle reconstruction and multi-task event reconstruction and classification from low-level detector information. AlphaHits-TrackFormer reconstructs the full charged-particle content of complete TrackML events without imposing a transverse-momentum selection on the target tracks. It processes both pixel and strip hits in events containing up to $1.5\times10^5$ hits under HL-LHC-like pileup conditions and assigns hits to individual reconstructed trajectories across its kinematic range. This includes the challenging low-$p_T$ population down to approximately $150\,\mathrm{MeV}$ in TrackML samples. It achieves a TrackML score of 0.959 over all reconstructable tracks, representing, to our knowledge, the highest reported TrackML score for full-detector, inclusive charged-particle reconstruction. Building on the TrackFormer architecture, AlphaHits is pretrained on charged-particle tracking and fine-tuned for multi-task reconstruction and classification. From a shared hit-level representation, it predicts tracks, jets, jet flavour, and event-level observables. AlphaHits achieves an AUC of 0.944 for Higgs event classification, with performance comparable to a Particle Transformer (ParT) baseline using fully reconstructed inputs. It also achieves performance comparable to Delphes-based baselines for jet reconstruction, jet flavour classification, and missing transverse momentum regression. These results highlight the potential for scalable full-event tracking and multi-task learning directly from detector hits, representing a promising step towards transferable hit-level representations and general-purpose foundation models for LHC physics.
Quantum Physics
Exceptional points in non-Markovian quantum-state diffusion: From coefficient dynamics to state signatures
An exceptional point of a memory-coefficient equation need not have the same signature in every quantum-state observable. We examine this relation within exact non-Markovian quantum-state diffusion (NMQSD), following the same operator closure from its coefficient dynamics to the non-Markovian master equation obtained by noise averaging. A finite homogeneous lift identifies the Jordan chains reached by the causal initialization, while state reconstruction determines which contributions survive in a specified readout. Two models expose the distinction. For common-bath qubits restricted to zero and one excitation, a noise-independent lowering-form $O$ operator gives a closed master equation. Every defective decaying mode transfers its full pole order to a suitable unconditional excitation–reference coherence. A known third-order configuration illustrates both a quadratic–exponential coherence and cancellation of its highest-order term in another quadrature. For a position-coupled oscillator in a fixed-photon-number optomechanical sector, the exact closure contains a scalar noise-history term. On the regular normalizable branch studied for bath decay rates below the oscillator frequency, two second-order coefficient blocks lead to a double pole in the conditional displacement response to the NMQSD noise history, at long preparation age and fixed lag. The purely algebraic coefficient tail cancels from the conditional Gaussian width. Complete noise averaging changes the master-equation coefficients and combines conditional width with the weighted spread of state centers. Together, the models show how causal preparation, state reconstruction, and averaging give coefficient exceptional points a precise, readout-dependent meaning.
Read the full abstract
Optimization and Control
Optimality Guarantees for Minimum Variance Subset Selection via Semidefinite Programming
Given a set of data points in Euclidean space and a prescribed cardinality $k$, the minimum variance $k$-subset problem asks for $k$ points minimizing their average squared distance from their centroid. We develop a semidefinite programming (SDP) framework for exact recovery and global optimization. First, we show that the natural SDP relaxation fails Slater's condition. Identifying the proper face containing its feasible region yields an equivalent, lower-dimensional, strictly feasible formulation that enjoys strong duality. A primal–dual characterization of optimality then gives geometric necessary and sufficient conditions under which the SDP relaxation is guaranteed to be tight. These conditions reveal how tightness depends on the internal geometry of the selected subset and on its separation from the remaining points. Second, we embed the SDP relaxation in a branch-and-cut algorithm combining first-order optimization, valid inequalities and dimension-reducing branching. Extensive experiments on public data sets, with up to 1,500 points and different values of $k$, show that the method strongly outperforms general-purpose exact solvers as the number of points and cardinality increase.
Read the full abstract
Machine Learning (Statistics)
MediPFN: Amortized In-Context Nuisance Learning and Inference for Causal Mediation Analysis
Causal mediation analysis decomposes a treatment effect into indirect pathways through mediators and direct pathways through other processes. Tabular foundation models perform classification and regression through in-context learning, providing estimates of conditional means and propensity scores used in causal mediation analysis. Constructing confidence intervals from these estimates requires accounting for uncertainty introduced by fitting and combining several nuisance functions. We introduce MediPFN, an amortized inference framework that adapts a pretrained tabular foundation model to nuisance prediction and combines its predictions through a cross-fitted influence-function estimator. A separate head learns quantiles of the estimator's standardized error from simulated tables with known indirect effects. Supervised fine-tuning, symmetrization, and independent calibration then produce a confidence interval. We establish coverage bounds for the resulting intervals under standard assumptions. Across experiments, MediPFN attains coverage closer to the nominal level than several comparators. It also reduces the computational cost of uncertainty quantification, running 59.1 times faster than bootstrap in matched comparisons.
Read the full abstract
Machine Learning
Wrong the Same Way: Cross-Prompt Sharing Both Suppresses and Amplifies Reward Hacking in RLVR
Two accounts of how reinforcement learning with verifiable rewards (RLVR) interacts with an imperfect verifier coexist: in one, reward hacking is inherited from the base model; in the other, RL amplifies it. We study both in a controlled environment small enough to compute the entire response distribution exactly. We vary the universality of a leak, or the fraction of prompts that accept the same wrong response, and the verifier's static false-positive rate. At its KL-regularized optimum, RLVR acts as a filter that reweights the accepted set without altering its composition, so the hack rate is inherited exactly. Taking that as a null model, a known decomposition writes the drift from it as a sum over prompts of covariances between reward and the empirical tangent kernel. The terms from other prompts vanish without parameter sharing. With sharing, every prompt that accepts a wrong answer pushes it up and every prompt that rejects it pushes it down through the baseline, in proportion to the model's probability of producing it there. A shared-parameter policy thus drives a leak that few prompts accept below the base rate and amplifies one that most prompts accept. At every convergent learning rate under Adam, SGD, and sampled GRPO, hacking ends higher at high universality than at low, but whether either end crosses the base rate depends on the step size. Independent per-prompt copies remove the dependence. The direction of this push depends on neither the verifier's error rate nor universality alone, but on how much of the model's kernel-weighted probability of a wrong answer falls on prompts that accept it, which frozen base model controls support. Verifier fixes should target the mistakes the model repeats across prompts: two patches that remove equally many errors differ by up to 38 points of true accuracy. We propose this alignment as a hypothesis for why earlier studies disagree.
Read the full abstract
Artificial Intelligence
Reliability of Claude Code as an Agentic AI Grading System for Handwritten University Exams
This study examines the reliability of Claude Code, an agentic deployment of the Claude Opus 4.6 large language model, as a grading system for handwritten university exams. Four research questions addressed agreement with expert human grading on total scores (RQ1), agreement across five question types from multiple choice to evaluative judgement (RQ2), consistency across two independent grading sessions one week apart (RQ3), and the accuracy of AI-generated feedback (RQ4). Claude Code graded the 20-mark mid-term papers of 60 consenting students in a Saudi computer graphics course in a single agentic session, guided by three rubric files. Reliability was estimated on disjoint samples: 20 papers for RQ1, RQ2 and RQ4, and 15 papers for RQ3. Analyses used Pearson correlation, mean absolute error, Bland-Altman analysis, weighted Cohen kappa, intraclass correlation, and a four-category coding scheme separating diagnostic from confirmation feedback. The author, as the sole subject-matter expert for the course, provided the human benchmark, so the design assesses agreement with a domain-expert reference rather than inter-rater reliability. AI grading agreed strongly with human grading on totals (r = 0.97; MAE = 0.85 marks) with a small but significant AI-positive bias of 0.68 marks, showed excellent test-retest consistency (ICC = 0.96), and produced accurate feedback (92.8% diagnostic; 97.8% confirmation). Drawing items were the outlier on three of four measures. Three failure modes recurred: keyword matching without conceptual depth, selection-state ambiguity on corrected multiple-choice answers, and region-of-attention drift on hand-drawn diagrams. The findings suggest that AI grading at this capability level can assist instructors on objective and short-form items under structural safeguards but is not yet ready to replace instructor judgement at boundary cases or on drawing items.
Read the full abstract
Digital Libraries
Latin American Studies since the 1980s: From Apocalyptic Dreams to Assimilationist Consolidation
We investigate how Latin American Studies (LAS) has evolved since 1982 in three leading LAS journals. Using network analysis, text mining, and bibliometrics, we trace transformations in disciplinary positioning, authorship, and citation patterns. Our main claim is that LAS has undergone an assimilationist consolidation. The field is now institutionally recognisable, with core LAS journals occupying central positions in disciplinary networks and acting as bridges across fields. Yet this consolidation has reinforced hierarchies privileging Anglophone academic norms. Journals based in Latin America, once central, have been marginalised. The share of Latin American authors has grown, but their profiles converge towards Northern standards.
Read the full abstract
Artificial Intelligence
Counting Under Disagreement: Dollar-Weighted Measurement, Selective Machine Labelling and Budgeted Human Review from Disagreeing AI Readers of Administrative Text
We study AI-assisted measurement when one text label applies to many administrative records, the target is a monetary total, and the human readings that anchor the machines disagree. The testbed is 121,070 grant-purpose texts from US tax filings (799,675 records, \$126.4B), labelled by a rule-based classifier and three language models, with a designed sample of 651 texts read independently by three people under one codebook. We state exact conditions for a fixed-predictor difference estimator and for a cross-fitted predictor whose folds are assigned to the whole frame before sampling, under Poisson sampling, and assess the implemented fixed-size procedures by simulation: 8 predictors whose own totals span \$11.2-15.0B agree to within \$0.4B once rectified. Two rules that accept machine labels at a stated dollar-weighted error, weighted conformal p-values under a dollar measure and Learn-then-Test on the design estimate of the error, are evaluated on nested held-out folds and on re-draws of the design with synthetic truths; their error control in deployment is not established, and the count-based rule accepts no text under a dollar-drawn sample. In a replay that charges every human label, a two-phase review policy whose calibrator is fitted on its own first phase reduces the RMSE of the total by 28% at a budget of 100 texts relative to dollar-proportional sampling, while pure Neyman allocation fails with an uncalibrated predictor. Independent readers shift the estimated total by \$3.93B, 2.2 times the within-reader sampling interval, and institution-level intervals under-cover at every size. Every headline result is re-run under each reader. The results identify both the benefits and the limits of using learned predictions to decide what a person should verify.
Read the full abstract
Networking
Collision Avoidance and Collision Detection Strategies to Counteract Large-Scale FMCW Automotive Radar Interference
Automotive radars operating in dense traffic environments may experience significant mutual interference, potentially degrading sensing reliability for advanced driver assistance systems (ADAS). This issue motivates the development of techniques to mitigate interference among multiple radar-equipped vehicles sharing the same spectrum. This paper investigates interference mitigation strategies for uncoordinated radar networks through system-level simulation. Two approaches are considered: frame-by-frame collision avoidance (FbF-CA), which exploits idle intervals to measure the perceived interference and possibly adapt the operating frequency, and frame-by-frame collision detection (FbF-CD), which estimates interference during transmission based on observed chirp collisions. An event-driven radar network simulator is used to model realistic highway traffic and radar interactions. Results show that collision avoidance provides higher robustness when a larger bandwidth is available, while collision detection is more effective under bandwidth-limited conditions. The analysis highlights the trade-off between collision avoidance and collision detection mitigation strategies and provides design insights for scalable automotive radar deployments.
Read the full abstract
Computer Vision
Before Words, Beyond Speech: Evaluating Nonverbal Social Reasoning in Early Childhood
Children's earliest years are a period of intense social learning, during which meaning is built through gaze, gesture, and physical contact. Yet, this critical aspect of social interaction remains under-evaluated in multimodal video understanding, as existing benchmarks prioritize the conversation-driven interactions of adults. We address this gap with \bench ($\textbf{N}$aturalistic $\textbf{E}$arly-childhood $\textbf{S}$ocial in$\textbf{T}$eractions), the first child-centric benchmark for nonverbal social reasoning. \bench comprises \totalvids manually annotated clips ($\sim$12s) curated for diversity across naturalistic settings, cultural contexts, and developmental stages. Beyond a lack of data, targeted model improvement in this domain is hindered by coarse evaluation schemas that conflate basic perception with higher-order reasoning. To resolve this, \bench introduces a compositional schema that builds atomic interaction cues into progressive levels of social reasoning: contextual, behavioral, interpersonal, and open-ended descriptions. Extensive analysis of state-of-the-art Vision-Language Models (VLMs) reveals a stark performance collapse: despite strong contextual recognition, models exhibit substantial gaps in decoding precise behavioral and interpersonal cues. We trace these failures to specific biases, including poor consequence grounding and temporal sensitivity. Finally, we ground \bench in real-world applications by evaluating VLMs using domain adaptation strategies. Ultimately, \bench provides a rigorous testbed for advancing socially grounded artificial intelligence in early childhood. Project page: https://laubravo.github.io/nest_website/.
Read the full abstract
Statistics Applications
Prediction Inference for Hybrid Models on Realized Volatility Forecasting
GARCH-LSTM hybrid models have become increasingly popular in the volatility forecasting literature. To date, these models have almost exclusively focused on point estimation and accuracy measures such as RMSE and MAPE, while failing to assess the uncertainty of these point estimates. We approach this gap in two ways. First, we extend the existing GARCH-LSTM model by modeling the conditional mean with an ARMA structure. Second, we develop a forward bootstrap algorithm to build multi-step-ahead prediction intervals for future realized volatility. In order to address the notorious undercoverage of prediction intervals, we implement two solutions. One draws bootstrap innovations from the empirical residual distribution to mimic the fat-tail property of financial series. The other utilizes principles from conformal prediction to systematically extend the prediction interval length. We evaluate the methods in simulations with heavy-tailed and time-varying GARCH processes and in rolling-window forecasts of four U.S. stock indices from 2020 to 2025. The ARMA-based hybrid gives the lowest point-forecast errors in most settings, and combining both remedies with the mean-based hybrid yields intervals closest to the nominal level among competing methods.
Read the full abstract
Human-Computer Interaction
Designing UIs with GenAI: Understanding the Use, Benefits, and Limitations of Prompt-to-Design Generators
The integration of generative AI (GenAI) into user interface (UI) design has led to prompt-to-design generators (P2dGens), tools that enable designers to create UI prototypes from textual prompts efficiently. Despite their growing adoption, little is known about how designers use these tools in everyday practice, their benefits, and limitations. We conducted semi-structured interviews with 17 UI designers to investigate P2dGens use, focusing on workflow integration, attitudes toward generator-assisted design, and the organizational climate. Findings show that designers primarily use P2dGens as ideation accelerators rather than end-to-end solutions, and they adopt a variety of individual and collaborative prompting strategies. Designers expressed complex attitudes, noting gains in efficiency and inspiration while also highlighting concerns about loss of personal creativity and design quality. We also observed emerging governance practices and varying attitudes within organizations toward P2dGens. These insights illuminate how P2dGens reshape UI design workflows and contribute to understanding human-AI collaboration in UI design.
Read the full abstract
Machine Learning (Statistics)
Near-Optimal Sample Complexity of Deterministic Agnostic PAC Learning
We study the problem of agnostic PAC learning under deterministic labels, where the examples are labeled by an arbitrary deterministic function, which need not belong to the concept class $C$. This model captures misspecification without stochastic label noise and lies between the realizable and fully agnostic settings. Prior work showed that, when $C$ has finite diameter $K$ and VC dimension $d$, deterministic labels can improve the sample complexity upper bound from the classical agnostic guarantee $O(d/\varepsilon^2)$ to $\widetilde{O}(K/\varepsilon)$, but left a large gap between upper and lower bounds. Our main result sharpens this picture. We prove that every class of VC dimension $d$ and finite diameter $K$ is learnable with sample complexity $\widetilde{O}\left( \frac{K^{1/3} \, d^{2/3}}{\varepsilon} \right)$, and we construct classes for which this bound is tight up to logarithmic factors. We also study the online analogue of this problem and show that the minimax regret is again partly controlled by the diameter of the class. Technically, our upper bounds are obtained via memorization-based learning algorithms that exploit the deterministic label structure.
Read the full abstract
Databases
ARoute: Task-Aligned Batch and Streaming Model Routing
Language models increasingly power data-intensive applications, from semantic data processing to question answering and agentic workflows. Using frontier models for every request is prohibitively expensive, while relying only on smaller models reduces accuracy. We study model routing under a user-specified cost budget, aiming to maximize accuracy under task-specific correctness criteria. Existing routers fall short in two ways: pretrained routers rely on generic preference signals that may not reflect a task's own correctness criteria or a new model pair, and model cascades run the weaker model first on every request, adding sequential latency. We present ARoute, a task-aligned router that learns cost and accuracy signals to select a model for each request before inference. ARoute has two variants: ARoute-Batch operates on a batch of requests known in advance, routing requests with the greatest expected benefit to the stronger model subject to the budget, whereas ARoute-Stream processes requests online, using a threshold calibrated on a small validation set to satisfy the cost quota and dynamically adjusting it as the budget is consumed. For ARoute-Stream, we provide a finite-sample guarantee on expected normalized spending. We evaluate ARoute on MMLU-Pro, LiveCodeBench, SuperGPQA, and two value-filling tasks built from SWAN semantic query benchmark against other state-of-the-art approaches. At matched cost, ARoute-Batch achieves the highest mean accuracy in all 15 task-cost settings, improving over the strongest baseline by 4.1 percentage points on average and by up to 11.9 points. ARoute also completes a batch up to 2.8x faster than cascades.
Read the full abstract
Geometric Topology
Thompson indices of knots and the invertibility problem
Jones showed that Thompson's group $F$ produces all unoriented links, and introduced its oriented subgroup $\vec{F}$, which produces all oriented links; the same holds for the Brown-Thompson group $F_3$ and its oriented subgroup $\vec{F}_3$. The corresponding Thompson indices measure the complexity of a knot in these groups, much as the braid index does for braid groups. In this article we prove that the span of the Jones polynomial plus $1$ is a lower bound for the $F$- and $\vec{F}$-index, and we give lower bounds for the $\vec{F}$- and $\vec{F}_3$-index in terms of the degrees of the HOMFLY-PT polynomial. We then compute the four indices (the $F$-, $F_3$-, $\vec{F}$- and $\vec{F}_3$-index) for the prime knots with at most 13 crossings, by an exhaustive enumeration of tree diagrams (up to 13 leaves for $F$, 17 for $\vec{F}$, with the HOMFLY-PT polynomials of all the diagrams with 19 leaves, and 9 and 11 for $F_3$ and $\vec{F}_3$) completed by targeted searches; every value is rigorously certified, by exact computations and by verified computations of hyperbolic structures in interval arithmetic. The $F$-index is determined for all knots with at most 11 crossings, the $\vec{F}$-index for all knots with at most 8 crossings, and the $F_3$- and $\vec{F}_3$-index for all knots with at most 12 crossings. The main outcome of these computations is that the oriented indices are sensitive to the orientation: we find hundreds of non-invertible knots $K$ whose reverse $K^r$ has a different $\vec{F}$- or $\vec{F}_3$-index, the smallest being $8_{17}$ (with $\vec{F}_3$-index $7$ and $9$). The $\vec{F}$-index also distinguishes the Conway knot from its mutant, the Kinoshita-Terasaka knot.
Read the full abstract
Computation and Language
HiCAP: Hierarchical Cultural Adaptation via Personality for Dialogue Generation
Cultural adaptation in dialogue systems refers to the ability to generate responses that conform to linguistic norms and social expectations across different cultures. However, existing dialogue generation approaches pay limited attention to the role of personality in shaping cultural expression and communication style. We therefore propose HiCAP, a novel framework that integrates personality traits into culturally adaptive dialogue generation. HiCAP employs a personality-conditioned control mechanism that hierarchically modulates response style by first aligning it with the cultural tone and then refining linguistic expression and content according to individual personality attributes. To embed these factors, the model combines supervised fine-tuning and instruction-based tuning, supported by dynamic instruction templates and multi-agent demonstrations to capture diverse combinations of culture and personality. Experiments on CulturalPersonas and CPED show that HiCAP achieves the strongest overall performance across various dialogue generation metrics, demonstrating effective cultural adaptability and personality modelling.
Read the full abstract
Statistical Methodology
Ratio-of-mediator-probability weighting for causal mediation analysis: estimation, extensions, and practical guidance
Ratio-of-mediator-probability weighting (RMPW) estimates natural direct and indirect effects by reweighting mediator distributions across treatment conditions rather than modeling the outcome, thereby accommodating treatment-by-mediator interaction. This article brings existing RMPW methods together within a common estimation framework, incorporates a heteroskedastic normal model for continuous mediators, and extends two-step variance estimation to multiple parametric weight components. The framework is illustrated using a national health survey to decompose disparities in cigarette smoking by sexual-minority status and a randomized welfare-to-work experiment to examine pathways from program assignment through employment to maternal depressive symptoms. Targeted simulations evaluate model specification, alternative weighting strategies, treatment-selection adjustment, weight stability, variance estimation, and continuous treatments. Parametric RMPW performed well when models were adequately specified and weights were stable, but correct specification alone did not prevent unstable estimation from concentrated weights. Stratification-based weighting provided some robustness to model misspecification. Standard errors that treated estimated weights as fixed were unreliable, whereas two-step and bootstrap inference performed well under appropriate conditions. The results provide practical guidance for selecting weighting strategies, diagnosing instability, and conducting inference. The methods are implemented in the community-contributed Stata command mmwsmed.
Read the full abstract
Numerical Analysis
Patch LegONet: Resolving Localized High-Frequency PDE Dynamics with Low-Frequency Operator Blocks
Localized interfaces, layers, and shocks can require high spatial resolution even when most of a partial differential equation (PDE) solution remains smooth. In global spectral methods, resolving these features requires increasing the approximation order across the entire domain. LegONet builds PDE solvers from reusable mechanism blocks, but its global coefficient representation faces the same limitation. We introduce Patch LegONet, which reuses pretrained mechanism blocks on smaller physical patches to resolve localized fine scales. The solution consists of a global background and local corrections. Mapping a fixed reference representation onto smaller patches increases local resolution without enlarging the blocks. The corrections join smoothly to the global background at patch boundaries. Differential operators are rescaled analytically to account for each patch's physical size. Nonlinear mechanisms are evaluated on the combined solution to include interactions between the background and local corrections. Patches can move, merge, or be removed during rollout while the pretrained patch blocks remain frozen. Six experiments on Allen-Cahn and viscous Burgers equations assess static, multiple, and moving patches in one to three dimensions. At comparable numbers of degrees of freedom, Patch LegONet consistently lowers the final relative solution error compared with global refinement. The reduction is approximately one order of magnitude in most tests. These results extend mechanism reuse across spatial scales and provide a practical route to local refinement in learned PDE solvers.
Read the full abstract
Robotics
Ground Reaction Force-Aware Model Predictive Control for Underactuated Upright Physical Assistance
Many rehabilitation robots execute prescribed motions using reactive tracking controllers that do not explicitly incorporate the user's instantaneous human-ground interaction. This paper presents Ground Reaction Force-Aware Model Predictive Control (GRF-MPC), a framework for upright physical assistance. By integrating live GRF and center of pressure (CoP) measurements directly into a human centroidal dynamics model, GRF-MPC formulates a real-time quadratic program (QP) that optimizes assistive forces and moments over a receding horizon. We deployed GRF-MPC on the Robotic Upright Stand Trainer (RobUST) in an underactuated 4-cable configuration and evaluated it against a fully actuated 8-cable Cartesian impedance baseline in nearly 300 trials across six healthy participants. Across tasks spanning impulsive perturbations, lunge trajectories, and voluntary limits of stability, GRF-MPC reduced peak planar center-of-mass (CoM) error by 19.6% during perturbations, lowered trajectory tracking error by 24.1%, required 28.7% less net mechanical work, and permitted a 45.9% greater normalized voluntary excursion area. The underlying QP solves in under 1 ms, comfortably enabling 100 Hz replanning and demonstrating that incorporating live human-ground interaction into predictive control yields stronger, more transparent physical assistance with half the actuators.
Read the full abstract
Machine Learning
Audited Gradients, Anchored Policies: Model-Based Reinforcement Learning for Efficient Flow Control
Feedback controllers for flows are expensive to design, since each candidate must be evaluated with a high-fidelity solver. Differentiable surrogates reduce that cost, but they also introduce a failure mode that prediction error cannot detect. Policy optimization uses the derivative of the objective with respect to the action, which we call the action gradient and which can be computed either through the surrogate or through the high-fidelity solver. Optimization through the surrogate uses only its action gradient, and a surrogate can follow the flow trajectory closely while misestimating this derivative. The surrogate action gradient is compared against a central finite-difference estimate from the high-fidelity solver on held-out states, under acceptance criteria fixed before the comparison, and we call that comparison an audit. A low-dimensional feedback law is then fitted to the surrogate action gradient when the surrogate passes the audit, and to the solver estimate when it does not. This law initializes a neural policy, and a quadratic penalty holds the policy near it during optimization. The proposed approach is validated on four test cases. The resulting controllers outperform or match a model-free baseline, while using 3 to 156 times fewer high-fidelity solver interactions across the four test cases. One surrogate predicts the objective at correlation 0.9996 but its action gradient at only 0.459, so prediction accuracy does not identify a usable surrogate. Another surrogate, whose action gradient agrees with the solver estimate at correlation 0.996, makes the objective 0.52% worse when the policy is optimized through it without the anchor penalty, so an accurate gradient alone does not provide a working controller. Finally, a surrogate that passes the audit at one fixed action fails it around the actions the deployed controller applies, so a pass is local.
Read the full abstract
Systems and Control
Robust State-Parameter Estimation in Second-Order Thévenin Batteries
We propose a scheme for estimating the full states and parameters of a second-order Thévenin battery model, using distinct features of the measured current. An identification step, run during suitable rest intervals, identifies the resistances and capacitances of the two RC branches from the voltage relaxation by linear least squares, without knowledge of the open-circuit voltage (OCV) map. An online estimator then runs continuously: the polarization voltages are estimated by open-loop observers; the series resistance is updated at current jumps using least squares; the state of charge (SOC) is propagated by Coulomb counting using the current capacity estimate, and both estimates are updated periodically by a sliding-window regression. We show that the identification is exact under a mild rank condition, that the series resistance is recovered at the first current jump, and that the online state and capacity estimates converge exponentially under verifiable excitation conditions. We also establish robustness of the online estimator under bounded process disturbances and measurement noise.
Read the full abstract
Optimization and Control
RALES: A Risk-Averse Augmented Lagrangian Equality-Constrained SQP Method
This paper develops RALES, an augmented Lagrangian method for risk-averse optimization in function spaces with nonsmooth risk measures, nonlinear equality constraints, and convex geometric constraints. The method combines safeguarded risk and constraint multiplier updates with efficient sequential quadratic programming based on inexact subproblem solves. The analysis establishes asymptotic stationarity for nonconvex problems and gives conditions for feasibility of weak primal accumulation points. Under additional assumptions on subproblem accuracy, duality, and boundedness, it provides global primal optimality guarantees and weak convergence of the full stored and candidate multiplier sequences to a common dual maximizer. Dual convergence is strong when all penalty parameters remain bounded. Numerical experiments for a finitely sampled semilinear PDE control problem, with the Conditional Value-at-Risk and the Mean-Plus-Semideviation risk measures, illustrate the method's performance as the spatial mesh is refined and as the number of samples increases.
Read the full abstract
Computer Vision
Image Quality and the Reliability of Fabric-Density Measurement: Counting Threads from Smartphone Photos Through a Pick Glass
Thread density, the threads per inch along and across a woven fabric, is a basic specification that buyers check whenever fabric is sold. Without laboratory instruments, inspectors count it slowly, by eye, through a pick glass, a folding 10$\times$ magnifier with a one-inch window. A smartphone photo through that glass could make the check faster and leave a record. We judge such a photo by what it is used for, as in task-based image quality: it is good if the thread count from it is within $\pm$5% of the true count, a common tolerance in fabric specifications. We introduce PickGlass-TPI: 76 factory scenes of 44 fabric patches, each counted in an open annotation tool by a textile-trained annotator, whose counts are the reference. On it we benchmark frontier and open vision-language models, CNNs, counting specialists and a training-free spectral counter (PG-Lattice). No method agrees with the reference within the tolerance. The best methods are off by about a quarter: a small open model that we tuned only on synthetic images (Molmo2-4B, 25%) is level with the best frontier model, GPT-6.1-sol (27% when asked directly, 25% given four magnified close-ups of the window). Both err less than a non-expert annotator (37%), but only the tuned model also beats, on both error measures, a baseline that guesses each scene's density as the average of the others. PG-Lattice most often repeats its count within 5% across two photos of the same patch, but its misses, like GPT-6.1-sol's errors, are often a factor of two. Even the reference repeats a count across such re-shoots only to a median 9%, so one photo cannot yet be shown to meet $\pm$5%; the image quality that matters is reliability across re-shoots. Generic no-reference quality scores chosen before labeling do not rank PG-Lattice's error, and generative enhancement raises perceptual scores but does not reliably improve the count.
Read the full abstract
Machine Learning
XAI-Driven Confidence Estimation for OOD-Resilient Runtime Prediction
Machine learning-based resource management (RM) models degrade under out-of-distribution (OOD) inputs at runtime, yet existing approaches in the literature rely on a single global error margin that is overly conservative and fails to distinguish reliablepredictions from catastrophic failures. We propose Lightweight Runtime EXplainability (LEX), the first post-hoc XAI framework for input-specific runtime confidence estimation in ML-based RM. LEX clusters the embeddings of the training input space using a Gaussian Mixture Model and associates per-cluster prediction error bounds to input samples, enabling fine-grained runtime confidence signals without retraining. Evaluated on a state-of-the-art LSTM-based IPS forecasting model on Intel Alder Lake, across 34 benchmark applications, LEX achieves ~94% containment on unseen workloads and produces ~17% tighter per-input confidence bounds than global baselines. With its negligible runtime overhead, LEX advances the field by providing explainable and reliable input-specific confidence signals of ML predictions, thereby making ML-based resource management resilient to OOD at runtime.
Read the full abstract
Machine Learning
Structured Kernel Surrogate Models for Multimodal Attribution
We study the problem of cross-sample modality attribution, which involves measuring how one modality in one sample affects the learning outcome in other multimodal samples. This problem is crucial in a multimodal system because a data modality (such as a text description) may be redundant for one sample (a self-explanatory image), but the same data modality would otherwise be uniquely informative for another sample, or informative when combined with other modalities. Prior work focuses on influence estimation at the level of one sample, and it remains an open question how one should capture the interactions across both samples and modalities. To capture scenarios like the above, we propose structured kernel surrogate models, which are surrogate mappings from both samples and modalities to the prediction outcome when they are trained together. We design an algorithm, SKS, which applies kernel ridge regression with a structured, compositional kernel on the selection states of individual samples and modalities. Specifically, the kernel function multiplies kernel-based similarity scores between every pair of data modalities for one sample and a pair of two samples. Additionally, we design a gradient-based estimation algorithm to efficiently estimate the structured kernel surrogate model. We evaluate our approach on two multimodal benchmarks, spanning a range of data modalities including satellite images, road networks, text, audio, and tabular data. We find that SKS achieves $57.8$% higher correlation with the true prediction outcomes than existing methods. In downstream data selection, SKS improves the average accuracy by $1.93$% over baselines. Models with data selected by SKS have lower Hessian traces, indicating a flatter loss surface. Using SKS to select satellite images identifies road segments, where accident scenes commonly occur, as opposed to background images.
Read the full abstract
General Relativity and Quantum Cosmology
Eccentricity greatly enhances the detectability of line-of-sight acceleration in gravitational wave signals
Compact object binaries formed dynamically are characterized by a tail of eccentric sources and by the potential presence of environmental perturbations. Here we show that the presence of eccentricity in a gravitational-wave (GW) signal strongly enhances the sensitivity to environmental effects, focusing in particular on line-of-sight acceleration (LSA). We demonstrate that this enhancement is recovered in Fisher-analysis forecasts and, crucially, persists in full Bayesian parameter estimation using state-of-the-art eccentric waveform models. The key insight is that the signal should be analyzed from the moment the first detectable eccentric harmonic enters the detector band, rather than from the customary entry of the $n=2$ harmonic: this allows to probe the early inspiral and thus a longer signal, over which the GW dephasing introduced by LSA becomes substantially larger. We show that with this approach constraints on LSA can increase by more than an order of magnitude between eccentricities at 20~Hz $e_{20}=0$ and $e_{20}\sim0.1$, for signals with sufficient signal-to-noise-ratio (SNR), while we predict a factor $5$ amplification even for low-SNR ones. Fully exploiting this effect necessitates the continued improvement of efficient eccentric waveform models and likelihood evaluations, enabling the efficient analysis of signals lasting thousands of seconds and containing a large number of harmonics.
Read the full abstract
Artificial Intelligence
From Process Models to Composable Agentic Workflows: Combining BPMN, Agents and Enterprise Knowledge
LLM-based agents can reason, use tools and collaborate. However, coordinating increasingly complex agentic applications becomes difficult to understand, monitor and improve. Decisions distributed across prompts, tool calls and application logic make workflows implicit. We introduce **Recursive Capability Composition**, enabling agents to execute business processes and expose them as reusable Agentic Tools. This paper presents a BPMN-based approach to Agentic Workflows, combining explicit orchestration with autonomous reasoning. SemTalk BPMN models are transformed into executable workflows using a LangGraph/Pregel runtime. Agentic Tasks introduce LLM reasoning at selected activities, while conventional BPMN controls process flows, deterministic decisions, system interactions and human approvals. Agentic Participants provide reusable interfaces for agent behavior or other executable processes. BPMN orchestration makes multi-agent systems more understandable and manageable. Explicit process models provide developers and operators with a shared view of workflow structures and execution progress, facilitating debugging, monitoring and optimization. Enterprise agents also need organization-specific context. Process Models and Knowledge Graphs describe activities, participants, information objects, decisions and relationships between processes. They provide structured knowledge of how an organization operates, supporting informed agent decisions and actions. Our contributions are threefold: 1. **Model-Driven Execution:** Transforming BPMN models into executable Agentic Workflows. 2. **Recursive Capability Composition:** Exposing agents and executable business processes as reusable tools for other agents. 3. **Enterprise Context:** Using Process Models and Ontologies to provide organization-specific knowledge for agent reasoning.
Read the full abstract
Astrophysics of Galaxies
TITAN DR1: The Type Ia Supernova Trove from ATLAS in the Nearby Universe - Host Galaxies I: Associations, Redshifts and Derived Properties
We present the host-galaxy catalogue for TITAN DR1, a sample of 8378 spectroscopically confirmed Type Ia supernovae (SNe~Ia) at $z \lesssim 0.15$ – the largest homogeneous low-redshift SN Ia sample assembled to date. We describe the association of each SN with its host, the compilation of redshifts, measurements of uniform far-ultraviolet to mid-infrared (FUV–MIR) photometry, and the derivation of host properties through spectral energy distribution (SED) fitting. Combining four host-association methods with manual vetting, we securely identify hosts for 7856 SNe~Ia (93.8%). We compile host-galaxy spectroscopic redshifts for 5522 objects and provide CMB-frame and peculiar-velocity-corrected redshifts. \rev{PSF-matched} photometry is measured for 7568 host galaxies, including both global fluxes and local optical-only fluxes within a fixed 1.5~kpc aperture centred on the SN; Bagpipes returns successful global fits for \rev{all} hosts and local fits for 7329. Stellar mass and rest-frame $g-z$ colour are robustly measured from optical data alone, whereas reliable estimates of star-formation rate, dust attenuation, and mass-weighted age require broader wavelength coverage to break age–dust degeneracies\rev{, and are reported for 5827 hosts}; in our redshift range, MIR data suffice and far-IR coverage is not required. The hosts span a representative low-redshift galaxy population, with a median stellar mass of $\log_{10}(M_*/M_\odot)=10.22$ (\rev{60%} high-mass; $\log_{10}(M_*/M_\odot)>10$), \rev{71%} star-forming ($\log_{10}(\mathrm{sSFR}/\mathrm{yr}^{-1}) > -10.5$), and an approximately exponential $A_V$ distribution. This catalogue provides the foundation for forthcoming analyses of the luminosity step and the broader role of galaxy environment in SN~Ia standardization and cosmology.
Read the full abstract
General Economics
Taste or Strategy? Nationalistic Bias and Competitor Penalties in Figure Skating Judging
Many decisions rest on expert panels whose members have ties to some candidates, and favoritism is hard to tell from shared taste or better information. Using a rich dataset from figure skating, we find evidence of a strategic compatriot premium and a targeted penalty on the compatriot's close rivals. The premium falls along a randomly drawn start order and, before 2022/23, rises when close rivals have their own judges on the panel. The penalty grows with the compatriot's placement chances and fades once she is out of contention. A counterfactual analysis shows that these biases can shape close competitions.
Read the full abstract
Geometric Topology
Vanishing Theorem of Pin(2)-equivariant Bauer-Furuta Invariants as Stable Sections
We develop a viewpoint of Pin(2)-equivariant Bauer-Furuta invariant: It can be converted to a section over a finite dimensional Pin(2)-orbit, and it takes values in ordinary stable homotopy class. This viewpoint reveals a topological limitation of the invariant: If the fiber's connectivity exceeds the dimension of the base space, then all sections are homotopic to the trivial section. We use it to recover Lin-Mukherjee vanishing theorem. Furthermore, the family Seiberg-Witten invariant can be obtained from the pullback of such sections to $S^1$-orbit. We use this fact to prove a vanishing theorem of the Pin(2)-equivariant family Bauer-Furuta invariant after two stabilizations.