Papers

Backtest auditing is a calibration problem: high flaw recall is not useful when the model falsely flags matched clean strategies. We build a 96-item paired benchmark in which every flawed backtest has a clean control that holds strategy, dates, code style, labels, and reporting scaffold fixed while changing one methodology detail. A deterministic scorer separates flaw recall, clean-control false positives, evidence localization, and fix relevance. Over 1440 cached audits from four text endpoints, the primary DeepSeek auditor reaches 100.0% closed and clean-aware code recall, but open prompts over-flag 93.8% of clean code controls, and clean-aware all-three specificity is 87.5% even where recall saturates. A clean-aware warning drops DeepSeek code false positives from 20.8% (95% CI 11.7–34.3) to 0.0% (0.0–7.4) at unchanged recall, while the budget anchor still flags 38/48 clean controls under the same prompt. Reporting recall alone would rank three of these four models identically; reporting the clean-control rate separates them by 79 points.

For every \(n\geq2\), we prove that the complement in \(\C^n\) of a periodic closed set with compact holomorphically convex quotient in \((\C^*)^n\) is Oka. The proof combines periodic divergence-free transport with Fatou–Bieberbach basins. Applications include complements of convex tubes \(\R^n+\ii B\), products of closed annuli, and, more generally, products of planar compact sets whose complements have at most one bounded component. In particular, \(\C^2\setminus\R^2\) is Oka.

In this paper, we complete the characterization of the boundedness of higher order commutators of pointwise multiplication and a large class of singular integral operators, when the corresponding kernels satisfy certain non-degenerate conditions. This was known with the restriction that the pointwise multiplier is real-valued. We finally remove this restriction by proposing a novel complex median method. We also showcase the flexibility of this method by proving the necessity of the rectangular product BMO for the bi-commutators.

The integration of model-based and data-driven paradigms provides a powerful framework for fault diagnosis by combining the interpretability of analytical redundancy relations, i.e., input-output relations that are used as diagnosis indicators in model-based diagnosis, with the adaptability of learning techniques. DT4X is a recent diagnosis algorithm that uses symbolic regression to generate multivariate relations leveraging some properties of analytical redundancy relations and uses them as split functions in a decision tree. However, its symbolic regression procedure optimizes only the separation between two selected classes at each node, often fragmenting the remaining classes and degrading both interpretability and diagnosis performance. This paper introduces DT4X+, an enhanced version of DT4X that modifies the construction of training sets and the symbolic-regression loss so that expressions separate the target classes while preserving the coherence of non-target classes. The resulting relations become fully consistent with ARR properties and lead to more informative splits, improved robustness, and better performance on dynamic-system datasets. Experiments conducted on several benchmark systems demonstrate the benefits of this enhanced formulation.

We propose LAYERSCOPE, a label-free, layerwise framework that aims to characterize a model's learned representations in video and multimodal settings. Evaluating downstream performance using representations from final or intermediate layers typically requires large amounts of labeled data, repeated task-specific evaluations, and substantial computation. To address these limitations, LAYERSCOPE uses local, global, distributional, and correspondence-based geometric metrics to compare layerwise representation structure within and across models without requiring task-specific labels. We evaluate seven architecturally diverse models across video and multimodal classification, clustering, and text-to-video retrieval tasks from MVEB/MVEB+. We find that intermediate-layer representations can outperform final-layer and model-default outputs. We also find that no single geometric metric consistently predicts downstream performance, but note that distinct layerwise geometric signatures emerge across model families. LID shows task-dependent relationships with performance, while RankMe provides the strongest measure for classification and clustering, but is not a universal layer selector. We also find that pairing-aware metrics explain retrieval better than distributional distances alone. LAYERSCOPE therefore offers a framework for comparing representations across models and layers, enabling a more systematic evaluation in video and multimodal settings.

The job shop scheduling problem is a challenging combinatorial optimization problem, and recent reinforcement learning approaches using graph neural networks have shown promise for learning scheduling policies directly from problem instances. However, training on large instances remains computationally expensive, and generalization across instance sizes remains challenging. This paper studies curriculum learning for graph neural network-based reinforcement learning in the job shop scheduling problem by comparing it with single-size training across three target sizes: 20 x 20, 25 x 25, and 30 x 30. In the curriculum setting, the policy is first trained on smaller instances and then progressively adapted to larger target sizes, allowing scheduling behavior learned in earlier stages to support learning on larger instances. Models are evaluated on unseen instances from 8 x 8 to 30 x 30 using the optimality gap, considering both generalization across all evaluation sizes and specialization on the target size. Results show that curriculum learning consistently reduces wall-clock training time, with larger benefits as the target size increases. The strongest advantage is observed at 30 x 30, where curriculum learning reduces the mean optimality gap across all evaluation sizes by approximately 8.1 percentage points, reduces the target-size mean optimality gap by approximately 8.6 percentage points, and saves approximately 50 hours of training time.

We introduce the Max-Min Covering Location Blocker Problem, a bilevel optimization problem in which a leader blocks a minimum-cost set of candidate locations so that the optimal coverage of a budget-constrained follower does not exceed a prescribed target. Each customer's coverage is determined by the least favorable open facility that can serve it. The induced follower coverage set function is nonmonotone and nonsubmodular, so the validity of interdiction cuts cannot be inherited from existing frameworks. We develop an exact nested decomposition that exploits the max-min coverage structure at both levels. At the outer level, we establish the validity of interdiction cuts from this structure and strengthen them with facility-specific coefficients bounding the coverage lost when a critical facility is removed. We solve the NP-hard follower problem by branch-and-Benders-cut after projecting out the customer coverage variables, and characterize an optimal dual solution of the separation problem in closed form, generating Benders optimality cuts in linear time without solving a linear program. Computational experiments on benchmark instances show that both ingredients substantially improve the performance of the method. With a moderate number of candidate locations, instances with 300,000 customers are solved to optimality within one hour. A comparison with two general-purpose bilevel solvers from the literature shows the proposed method to be one to three orders of magnitude faster, and to solve to optimality instances that neither of them closes. The model applies wherever the planner cannot control which facility serves a customer, as in public automated external defibrillator (AED) networks. A case study on a network using real AED data from Virginia Beach illustrates the model.

Digital zoom transitions between dual cameras often exhibit conspicuous discontinuities in geometric structure and chromatic consistency, degrading the user experience. While recent dual-camera smooth zoom (DCSZ) methods attempt to mitigate this by fine-tuning frame interpolation (FI) models on DCSZ data, they struggle with the large cross-view disparities and complex geometric transformations. Considering that the generative prior of diffusion models is suitable for addressing this problem, we explore their application to DCSZ. However, naively applying existing diffusion-based FI models still yields low-fidelity transitions due to insufficient conditional guidance, high-frequency information loss during VAE encoding, as well as inadequate temporal consistency. To address this, we propose ZoomDiff, a high-fidelity diffusion model that leverages dual-camera inputs in both latent and pixel spaces for photo-realistic transitions. Specifically, we first strengthen dual-image conditional guidance during the multi-step denoising process to improve geometric consistency. Then we inject flow-aligned multi-scale features from the VAE encoder into the VAE decoder to recover high-frequency details, where flow-guided temporal consistency supervision are introduced to produce more smooth transitions. Extensive experiments on both synthetic and real-world datasets demonstrate that ZoomDiff outperforms state-of-the-art methods quantitatively and qualitatively. Project page: https://jiayi-hit.github.io/ZoomDiff.github.io/.

We study the distributed average consensus problem under persistent link-level disturbances modeled as a martingale difference sequence with uniformly bounded conditional second moments. Under such disturbances, the standard stochastic-approximation-based linear iteration with diminishing stepsizes drives the network to consensus on an unbiased random variable with non-vanishing variance instead of the exact initial average. To understand and resolve this limitation, we develop an anchoring-based mechanism derived from a decentralized gradient descent formulation and study the effect of incorporating a decaying anchoring term that continuously pulls each agent state toward its initial value. This perspective provides an intuitive interpretation of how state anchoring counteracts disturbance accumulation. Under standard summability conditions, we prove that the resulting algorithm achieves exact average consensus almost surely. Furthermore, this decentralized gradient perspective offers a unifying framework for several related methods and an interpretable design principle for exact average consensus under persistent disturbances.

Conventional image registration algorithms are robust to domain shifts and achieve low errors, but they are slow and computationally expensive. Deep-learning methods are efficient at inference-time, but face challenges in out-of-domain samples. We propose Recursive Uncertainty-Gated Image Registration (RUGI), an algorithm for iteratively refining deformation fields predicted by learning-based registration models. At each iteration, the registration model predicts an incremental deformation, and a gating map modulates the update. Refinements are hence concentrated in regions that remain difficult to register. We explore two gating strategies: a learned uncertainty-based approach and an image residual error approach. We evaluate RUGI on cardiac MRI and echocardiography datasets and show consistent improvements over single-step inference. Ablation experiments demonstrate that iterative refinement alone improves registration, but informative spatial gating provides a significant additional benefit. The error-gated variant of RUGI can also be applied directly to existing pretrained models; applied to VoxelMorph, TransMorph, and CycleMorph, it yields MSE reductions of 27-37% with no modification to the original training procedure. The improvements in registration performance are reflected in decreased errors in ejection fraction estimation relative to ground truths. These results demonstrate that spatially selective iterative refinement provides an effective strategy to improve registration accuracy at inference-time.

Evaluating open-ended text generation involves understanding how different properties of a continuation relate to its perceived quality. We present a reference-based framework for examining coherence and diversity through three perspectives: aligning their evolution with human trajectories, comparing their summaries with a human continuation of the same prompt, and estimating their likelihood under a human reference distribution. Experiments with human quality ratings suggest that diversity-based alignment and mean-based comparisons capture quality-related variation, although the comparisons do not establish a predictive advantage for temporal alignment over simpler baselines. Reference likelihood also shows positive associations with ratings, with results varying across reference configurations and scoring horizons. Together, these analyses provide a structured way to examine how measured coherence and diversity relate to human judgments, while distinguishing similarity to human references from quality itself. Code and analysis resources are available at https://github.com/EstebanGarces/likely_human.

Aims. We report on the first X-ray detection of the symbiotic star CN Cha with eROSITA (extended Roentgen Survey with an Imaging Telescope Array) onboard Spektrum-Roentgen Gamma (Spektr-RG). Methods. We analyse the eROSITA observations from eRASS2 to eRASS5 individually, as well as the combined eRASS:5 data of all five surveys in order to investigate the X-ray spectral and timing properties of CN Cha throughout the eRASS. Results. Although CN Cha was not detected during the first eROSITA All-Sky Survey (eRASS1), it was detected for the first time in the second eRASS2. Our analysis shows that the X-ray emission of CN Cha gradually increased from eRASS2 to eRASS5. The X-ray spectrum of the source, with the main emission at <2.4 keV, allows us to classify it as a e̱ta}-type symbiotic star. However, a more detailed analysis suggests that there is very soft component, which may originate either from thermonuclear burning on the white dwarf, or from the wind collisions. Additional X-ray observations are required to better understand the origin and emission mechanisms of the source.

Segmentation-based multi-object tracking (MOT) with foundation video models such as SAM2 offers strong localization quality, yet remains fragile in crowded, real-world scenes. In detector-prompted SAM2 pipelines, failures typically arise at three stages of the object lifecycle: (i) erroneous or duplicate track initiation, (ii) memory drift during close interactions, and (iii) unreliable re-identification after long occlusions or re-entry. These errors corrupt object memory and accumulate over time, making long-horizon tracking unstable. In this paper, we reframe MOT as a lifecycle memory integrity problem. We present LiAM-SAM, a Lifecycle-Aware Memory (LiAM) framework with targeted mechanisms for each of the three failure modes. At track birth, to prevent faulty or duplicate initiations, we apply contrastive track initiation, which conditions each prompt on existing nearby tracked instances. To preserve memory integrity during strong interactions, we introduce motion- and geometry-grounded memory correction that resolves interaction confusions and suppresses drift. For reliable re-identification after disappearance, we maintain an adaptive context memory that promotes diverse and trustworthy references as long-term identity anchors. Finally, similarity aware spatial pruning optionally selects the memory tokens to retain at cross-attention time, improving efficiency with minimal accuracy loss. LiAM-SAM represents a modular, detector-agnostic, SAM2-based MOT system that achieves state-of-the-art HOTA and IDF1 on the evaluated benchmarks. In association-challenging environments, our ablations show that LiAM improves a detector+SAM2 baseline by +10.5 HOTA, +17.4 AssA, and reduces identity switches by 96%.

Steel surface defect detection is critical for automated industrial quality control but remains challenging due to subtle inter-class texture differences and pronounced class imbalance. We introduce TEEP-RCNN (Texture-Enhanced Edge-aware Perception Region-based CNN), a two-stage detector built on Faster R-CNN with a Feature Pyramid Network backbone and an improved Convolutional Block Attention Module (CBAM). Our CBAM adds dropout regularization in the channel attention MLP and batch normalization on the spatial attention branch, reducing co-adaptation and stabilizing gating logits. Training uses a differential learning rate protocol with cosine annealing warm-up, separating update rates for the pre-trained ResNet-101 backbone and the detection head. At inference, predictions are refined via Test-Time Augmentation fused with Weighted Box Fusion (WBF), improving localization stability on elongated and boundary-adjacent defects. On the NEU-DET benchmark across six defect categories, TEEP-RCNN achieves 73.3% mAP@50 and 37.9% mAP@50-95 in only 10 training epochs on a single GPU, competitive with YOLOv11m (76.2% mAP@50, 100 epochs) while outperforming it on the rolled-in-scale category under the COCO metric. Per-class analysis shows the spatial attention branch is most effective on elongated texture defects such as patches and scratches, while crazing remains an open challenge across both paradigms due to its distributed non-local texture structure.

Motivated by a recent article of R. Vanlaere (2026), we investigate the spectral asymptotics of a singular harmonic oscillator arising from the radial reduction of the magnetic Schrödinger operator on the unit disk in the presence of a constant magnetic field and an Aharonov-Bohm flux. The corresponding eigenvalue problem is equivalent to the study of the zeros of the Kummer confluent hypergeometric function with respect to its first parameter. Combining now standard semiclassical methods with the uniform asymptotic theory of Whittaker and confluent hypergeometric functions developed by Dunster (1989), Gabutti-Gatteschi (2001) and improved quite recently by Dunster (2026), we obtain a comprehensive description of the different spectral regimes. When the boundary lies in the classically forbidden region, we derive the exponentially small asymptotic correction to the eigenvalue caused by the tunneling effect. For energy levels above the tunneling regime, we derive a shifted Bohr-Sommerfeld asymptotics for highly excited states at a fixed magnetic field. Finally, for the transition regime where the energy level touches the boundary, we derive a three-term asymptotic expansion of the eigenvalue. In an appendix, we translate these spectral results into explicit asymptotic formulas for the zeros of the Kummer function.

We provide a general upper bound on multicolor ordered Ramsey numbers in terms of the interval chromatic number and the degeneracy of an ordered graph. We extend previous results by Conlon, Fox, Lee, and Sudakov (2017) by showing that for every $n$-vertex ordered graph $G^<$ with degeneracy $d\geq2$, and interval chromatic number $\chi$, its $q$-color ordered Ramsey number satisfies $r_<(G^<;q) \in n^{O(d^{q-1}\lceil \log{\chi}\rceil^{q-1})}$ for every $q \geq 2$. For fixed parameters $q,d,\chi$, the resulting estimate is polynomial in $n$. For triangle-free ordered graphs $G^<$, we also provide the stronger estimate $n^{O(q^2d {\lceil \log \chi \rceil}^{q-1})}$. It also follows from a recent result by Li (2026) that our upper bound is almost tight for ordered matchings.

Event-by-event mean-$p_T$ fluctuations probe initial-state fluctuations and their evolution through the dynamics of heavy-ion collisions. We study second-order mean-$p_T$ fluctuations in Au+Au collisions at $\sqrt{s_{\rm NN}}=3.0$–$19.6$ GeV using JAM2 in the RQMDv mean-field mode with the MH2 parameterization. The model qualitatively reproduces the measured identified-particle $p_T$ spectra, providing a single-particle baseline for the fluctuation analysis. The scaled fluctuation $k_2$ decreases with increasing $\langle N_{\rm part}\rangle$ and shows broad agreement with the available measurements at 7.7–19.6 GeV, whereas its calculated centrality dependence is stronger than that in the data at 3.0–4.5 GeV. For the combined proton-plus-antiproton sample, the unnormalized correlator $\langle c_2\rangle$ is positive and larger than that for charged pions. The charged-pion correlator $\langle c_2\rangle$ is negative or consistent with zero over most centrality intervals at 3.0 and 3.5 GeV, becomes weakly positive at 4.5 GeV, and remains positive at higher energies. Its energy evolution resembles the change in the reaction-plane elliptic flow, but this comparison does not establish a common microscopic origin. These calculations provide species-dependent predictions within a transport model without an explicit partonic stage. Isolating the contributions of mean fields, rescattering, and spectator interactions requires controlled variations of the transport dynamics.

Context. Chromospheric emission observed in the Ca II H and K resonance lines is generally positively correlated with photospheric magnetic fields. However, a fraction of strong Ca II H&K brightenings are found to occur over regions of low photospheric magnetic flux density. Aims. We aim to understand the origin of these chromospheric brightenings associated with weak fields in quiet Sun regions. Methods. We use Ca II H observations taken by the SuFI along with the photospheric magnetograms recorded by the IMaX, instruments onboard the first flight of the balloon-borne observatory Sunrise. We also use the atmospheric physical parameters inverted from IMaX data by the SPINOR code. Results. 217 Ca II H brightenings associated with weak fields are identified. 111 of these are located at the peripheries of strong field regions, in the magnetic canopies formed by the expansion of the kG flux tube. The other brightenings are associated with either weak unipolar magnetic patches, with interactions between weak mixed-polarity magnetic fields, or at locations with no clear magnetic features. For the latter two situations, the Ca II H intensity varies with a period of 3-5 mins, accompanied by variations in the line-of-sight velocity, the temperature, and the magnetic flux with a similar period. Conclusions. About half of the strong Ca II H brightenings associated with weak fields are located in magnetic canopies overlying very weak photospheric fields. About 10-12% are associated with magnetic bright points where the magnetic field is probably not spatially resolved. The remainder appears to be Ca II H grains, which are overwhelmingly associated with magnetic emergence or cancellation events. We conclude that although Ca II grains are mainly driven by (magneto-)acoustic shock waves, interactions between opposite-polarity magnetic fields also contribute to the formation of the brightest peaks.

Current and future large spectroscopic surveys are significantly enhancing the volume and resolution of quasar spectra that are observed, which requires the creation of efficient and precise automated techniques to detect absorption features. This study focuses on the detection of metal absorption features using a novel U-Net model on WEAVE-like mock spectra in the quasar rest-frame wavelength interval $1230\,\mathring{\mathrm{A}} \leq \lambda_{\rm RF} \leq 3095\,\mathring{\mathrm{A}}$. We test the network performance for absorption detection both on ideal data and after simulating the continuum fitting step as applied on real data. The performance of these architectures is evaluated by the completeness, purity, and F1 score reached in bins of signal-to-noise ($\mathrm{S}/ \mathrm{N}_\mathrm{line} $) for the absorption lines and with the absolute fractional flux error for the continuum. The ability to recover the correct line centers is also studied. The U-Net reaches scores of $\approx 90\%$ for all metrics (completeness, purity, and F1 score) at $\mathrm{S}/ \mathrm{N}_\mathrm{line} \approx 4$. All false positive detections with $\mathrm{S}/ \mathrm{N}_\mathrm{line} \geq 5$ fall in the tails of the broad Ly$\alpha$ absorbers distribution of damped Ly$\alpha$ systems. The combination of continuum fitting and line detections has negligible effects on the detection performance at $\mathrm{S}/ \mathrm{N}_\mathrm{line} \geq 4$. Our proposed U-Net architecture offers a competitive tool for the analysis of absorption lines in current and upcoming large spectroscopic surveys and is well-suited to the identification of any absorption line feature.

The Purcell effect has emerged as a powerful mechanism for controlling spontaneous emission and dissipation in cavity and nanophotonic systems; however, its realization in dissipative non-Hermitian hybrid platforms remains largely unexplored. In this work, we investigate the Purcell effect in a dissipatively coupled photon-magnon hybrid quantum system consisting of a yttrium iron garnet (YIG) thin film integrated with an inverted octa-ring resonator (IORR) in a planar geometry. To describe the underlying dissipative hybrid dynamics, we develop a quantum theoretical framework based on non-Hermitian coupled-mode theory combined with the input-output formalism. We also analyze the temporal decay dynamics of the photon-magnon system. Full-wave electromagnetic simulations demonstrate that the system can be engineered to operate in the level-attraction regime through dissipative photon-magnon coupling. By systematically tuning the magnon damping, we uncover the emergence of a Purcell regime in which cavity-photon dissipation is selectively enhanced through magnon-mediated loss channels. We further show that the saturation magnetization (Ms) provides an additional degree of control over the onset, tunability, and robustness of the Purcell enhancement. The combined modulation of magnon damping and Ms strongly influences the effective photon-magnon coupling, hybrid-mode evolution, and dissipation landscape, enabling precise control of hybrid quantum states. These findings establish a versatile strategy for engineering magnetization- and dissipation-controlled photon-magnon interactions in planar, chip-compatible architectures, opening new avenues for non-Hermitian cavity magnonics, tunable microwave dissipation engineering, and hybrid quantum information technologies.

We investigate strangeness production and transverse dynamics in heavy-ion collisions at $\sqrt{s_{\mathrm{NN}}}\approx 2.5-20~\mathrm{GeV}$ using the transport approach SMASH (Simulating Many Accelerated Strongly-interacting Hadrons), its extension with rope hadronization, and the SMASH+vHLLE hybrid approach. Results from the Pythia-based heavy-ion model Angantyr, with and without rope hadronization, are included for comparison. We study midrapidity particle yields and average transverse masses as functions of the number of wounded nucleons, as well as their energy dependence. For the $K^+/\pi^+$ ratio, SMASH+vHLLE overpredicts strangeness production at low energies but describes the higher-energy behavior reasonably well. SMASH+Ropes reproduces the ratio up to $\sqrt{s_{\mathrm{NN}}}\sim 10~\mathrm{GeV}$ but does not capture the turnover at higher energies. In contrast, the transverse-mass observables favor the hybrid approach, while the non-thermal models considered here do not generate sufficient collective transverse expansion. These results show that strangeness enhancement alone does not uniquely distinguish microscopic string interactions from a locally equilibrated medium. Simultaneously constraining strangeness production and transverse dynamics is therefore essential for disentangling thermal and non-thermal mechanisms in heavy-ion collisions.

We give a counterexample showing that an asymptotic mean value property for $p$-harmonic functions in $\mathbb R^d$ does not necessarily hold in the pointwise sense for $d\geq 3$ and $p>2$ near $2$. The proof uses a perturbation argument from $p=2$ and the Implicit Function Theorem to establish the counterexample. We also use a computer assisted proof to give an explicit range of values of $p$ for which the counterexample holds. Finally, we also establish a counterexample for the pointwise asymptotic variational mean value property.

Early reflections in spatial room impulse responses (SRIRs) measured using spherical microphone arrays (SMAs) play an important role in spatial audio analysis, rendering, and reverberation modeling. Their accurate localization and characterization can facilitate the analysis, processing, and manipulation of measured reverberation fields. This paper proposes a method for detecting and characterizing early reflections based on the Herglotz wavefunction formalism. The measured sound field is represented as a continuous superposition of incident plane waves, from which a localization function is derived in the spherical harmonic domain. An adaptive radial Gaussian fitting procedure is then used to estimate both the directions of arrival (DoAs) and the number of incident reflections within a single analysis frame. The proposed framework is evaluated using simulated and measured SRIRs and compared with conventional steered response power (SRP) and multiple signal classification (MUSIC) localization methods. The results demonstrate improved localization accuracy and more reliable detection of multiple simultaneous reflections, highlighting the potential of the proposed Herglotz-based formulation for the analysis of early reflections in SMA-measured SRIRs.

Observations suggest a connection between steady-wind mass loss and coronal X-ray activity in low-mass main-sequence stars. Interpreting this connection is challenging because the wind is controlled mainly by the large-scale open field, whereas X-ray emission traces heating in small-scale closed fields often unresolved in global wind models. Here we use Space Weather Modelling Framework-Alfvén-Wave Solar Model with global convective dynamo-generated magnetic maps and solar magnetograms. We model Alfvén-wave-heated winds for four solar-type stars plus the Sun, spanning rotation rates of $1.0$–$23.3$ times the solar rate, and magnetic field strengths of $6.0$–$1200$ G. Our models show that faster rotation yields a more tightly wound spiral, a larger Alfvén surface, higher terminal wind speeds, and a harsher wind-pressure environment for orbiting exoplanets, different from that of the present-day Sun. We estimate the mass- and angular-momentum-loss rates and find systematic differences from Zeeman-Doppler Imaging-based predictions. Building on the successful reproduction of X-ray coronae in our Paper~I, we obtain the first self-consistent activity–wind relation in a unified modelling framework: the mass-loss rate scales with the surface X-ray flux following a power law with an index of ${\sim}0.67$. We also re-examine magnetic braking via open-flux magnetisation, finding that the effective Alfvénic lever arm depends on the magnetisation parameter with a power-law index of ${\sim}0.35$. Finally, we quantify the stellar wind pressure at the orbits of several super-Earths. Together with our Paper~I, the series of results shows the distinct roles of multi-scale magnetic fields and provides physically grounded inputs for assessing habitable-zone space weather.

Quantum sensing protocols usually presuppose where and when each signal acts, yet this information is unavailable beforehand in experiments such as scattering and radar. We show that causal agnosticity forces a measurement scheme to be interferometric even when only discrete interactions are allowed. Describing the sensing events in the recently developed quantum state over spacetime formalism, we then prove that the Fisher information of an interferometer is bounded by the squared sum of the local generator norms, independently of how the events are ordered. Nonunitary and spatiotemporally correlated signals can be explained as well when their physical dilation has a spacetime state on the enlarged events. Notably, the bound holds without presupposing a background spacetime that orders the events whenever the full collection of target and auxiliary events has a spacetime state. We show that this bound yields the spacetime Heisenberg limit and characterize its saturation through marginals of the spacetime state `dressed' by the signal. The $N^2$ scaling of the Fisher information with $N$ sensing events arises from the coherence of a single reference system shared across spacetime, whereas independently repeated interferometry obeys the standard quantum limit.

We study the arithmetic of the realizations for hypergeometric motives attached to ${}_3F_2(1)$ hypergeometric series defined over number fields, focusing on the case where the étale realization is reducible and irregular. We then use motivic techniques to prove new evaluation formulas for certain ${}_3F_2(1)$ series, understood as a period of the hypergeometric motive. In addition, we provide examples which are closely related to the exact values of $L$-functions for modular forms in special cases.

Driving world-action models improve planning by coupling multimodal reasoning with future prediction, but their growing inference cost increasingly conflicts with the real-time latency requirements of vehicle control. Existing acceleration methods reduce tokens, layers, or sampling steps with policies selected prior to deployment, yet leave residual runtime variation largely unexploited after offline profiling and static scheduling on shared onboard compute. We observe that the largest admissible compute budget varies systematically with the residual runtime state, while recent realized latency provides a direct signal of the available compute slack. Motivated by this observation, we propose SlackDrive, a pre-inference compute allocator that reuses realized latency to select the compute budget of each control step before model execution. SlackDrive profiles the latency and planning utility of a small discrete budget set once, estimates online compute state from completed forwards, and selects the highest-utility budget predicted to remain within the admissible latency envelope, complementing existing profiling and resource scheduling while preserving the driving backbone and its compute actuator. On NAVSIM v2 with DriveDreamer-Policy, SlackDrive improves latency-constrained EPDMS by $21.7\%$ over the strongest baseline under a stringent latency regime, while the full-budget model and preconfigured token-pruning baselines exceed the admissible latency envelope under runtime contention.

Compressive sensing (CS) exploits intrinsic signal sparsity for efficient representation of high-dimensional data. However, conventional CS relies on fixed-length measurement representations and dense reconstruction operations, leaving communication and computational costs largely determined by system dimensions rather than signal sparsity. Here, we propose a neuromorphic CS framework centred on a spike-driven learned iterative shrinkage-thresholding algorithm (S-LISTA). By representing compressed measurements and reconstruction updates as sparse events, the framework links intrinsic signal sparsity to the spatiotemporal sparsity of spiking neural networks (SNNs), extending the benefits of sparsity across sensing, transmission and reconstruction. Communication and computational costs consequently depend on event activity, allowing sparse representations to translate into resource savings. Extensive experiments demonstrate substantial reductions in transmitted data volume and estimated reconstruction energy while maintaining competitive reconstruction and downstream task performance. The framework also exhibits robustness under challenging channel conditions. These findings establish neuromorphic CS as a promising paradigm for signal reconstruction in resource-constrained scenarios.

Classical continuous-time non-linear filtering fails in generic non-linear state spaces due to an infinite Lie algebraic derivative explosion ($\dim(\mathcal{E}) = \infty$), leading to filter divergence. To resolve this four-decade crisis, we introduce the SMG-Yau-Yau filter and lossless data assimilation framework built on Statistical Fiber Theory over an Orlicz manifold $\mathcal{M}$. By equipping $\mathcal{M}$ with a Riemannian submersion and an Ehresmann connection, the unconstrained Duncan-Mortensen-Zakai score velocity field is orthogonally decomposed into Statistically Verifiable Directions ($\text{SVD}\chi_f$) and Structural Internal Directions ($\text{SID}_f$). System non-linearities and unclosed Lie commutators are orthogonally quarantined in $\text{SID}_f$, protecting macroscopic base parameters from spatial derivative pollution while preserving total score variance energy. We unify the asymptotics through a Dual-Axis Collapse Mechanism: proving our model identically recovers classical Yau-Yau dynamics when $\dim(\mathcal{E}) < \infty$, while large-sample limits ($N \to \infty$) induce a thermodynamic quench that flattens infinite-dimensional geometry under $\dim(\mathcal{E}) = \infty$. Finally, we formulate the Active Acausal Tension (AAT) functional to monitor accumulated model misspecification online. Exceeding a topological capacity threshold triggers Gauge Symmetry Breaking (GSB), which dynamically expands base coordinates ($d \to d+1$) to ensure non-asymptotic stability and convergence to the exact state density.

Automated LOD3 building modeling typically relies on purpose-built geometric or learning-based pipelines, limiting flexibility across heterogeneous buildings and input evidence conditions. This study investigates whether Astra, a general-purpose multimodal foundation model, can address these limitations through zero-shot reconstruction of LOD3 building models within an agentic framework under bounded autonomy. AstraLOD3 combines multi-view images, calibrated cameras, and a filtered sparse SfM point cloud with a natural-language reconstruction specification, while the Astra agent dynamically selects and executes computational procedures using Python and Blender. Across 35 runs, including 24 benchmark buildings, AstraLOD3 achieved a mean FRDS of 0.9647 and geometric agreement comparable to that of previous purpose-built methods. Controlled ablations further revealed the effects of reconstruction guidance, evidence modalities, model configuration, and run-to-run variability. The results demonstrate that structured LOD3 reconstruction can be formulated as a constrained agentic process rather than as a fixed pipeline. Future work will investigate adaptive refinement, user-guided correction, task-specific specialization, and damage-aware reconstruction.