Papers

This work represents a crucial advancement in the development of a new class of turbulence closures conceived by Cimarelli et al. [1]. This new class of closures is based on the theoretical framework provided by the temporally filtered Navier-Stokes equations. This framework represents a unifying formalism for coarse-grained approaches to turbulence ranging from the statistical to the scale-resolving ones and provides exact equations relating turbulent stresses at different levels of coarsening. The latter have been used in Cimarelli et al. [1] as the theoretical foundation to derive a new class of closures based on the dynamic computation of the model coefficient appearing in the definition of eddy viscosity. Here, we extend this class of dynamic RANS models by deriving a new formulation based on the k-{\omega} model including an alternative version where the non-physical decay of free-stream turbulence reproduced by the transport equations of k and {\omega} is solved in an elegant way. A scale-resolving simulation campaign is performed and the analysis of the results reveals interesting features. Among others, the new dynamic k-{\omega} model demonstrates strong robustness to a coarsening of the temporal resolution and successfully captures complex flow phenomena, including laminar-to-turbulent transition and the sensitivity of boundary-layer separation and reattachment to free-stream turbulence levels prescribed at the boundaries.

GC–EI–MS is an important technique for analyzing volatile and semivolatile compounds in complex samples. However, conventional methods rely heavily on reference spectral library matching, limiting their ability to identify compounds absent from these libraries and to infer complete molecular structures directly from fragmentation information. Here, we present DiffGCMS, a spectrum-conditioned discrete graph diffusion model for de novo structure elucidation from GC–EI–MS, and further develop a framework that integrates DiffGCMS with second-stage reasoning by a large language model (LLM). In the first stage, DiffGCMS generates candidate molecular structures from input spectra; in the second stage, the LLM uses mass spectral information to validate, repair, and rerank the candidates and provides interpretable analysis of fragment-ion peaks. This framework can generate plausible molecular structures for compounds absent from reference spectral libraries and provide traceable evidence supporting its decisions. On a test set comprising 13,696 spectra from NIST 20, the generative model achieved Acc@1 and Acc@10 of 6.01% and 15.76%, respectively. On the test subset containing molecules with no more than 10 heavy atoms, LLM-assisted molecular graph repair and reranking increased Acc@1 from 21.28% to 21.95%, Acc@10 from 46.91% to 47.99%, and candidate validity from 91.04% to 100%. These results demonstrate that spectrum-aware postprocessing can correct errors produced by the generative model while providing auditable and traceable explanations for the final ranking.

Learning control from action-free recordings is challenging because intervention effects are unobserved and policies may exploit errors in reconstructed dynamics. We present a hierarchical model-based reinforcement learning framework that uses shared structure across related systems to learn system-specific control policies from action-free recordings. A hierarchical dynamical system reconstruction model captures shared dynamics and individual variation through low-dimensional embeddings. These embeddings are then reused to parameterize shared policy and value networks, linking differences in reconstructed dynamics to differences in control. Policies are trained entirely via simulation under an explicit intervention model with additive latent perturbations. Piecewise-linear recurrent neural networks enable mechanistic analyses of the controlled dynamics, while decoder-based constraints make the immediate effects of interventions interpretable in observation space and permit interventions on one modality while protecting another from direct manipulation. On Lorenz-63 and double-pendulum systems, hierarchical policies improve transfer over independently trained policies. On Lorenz-63, they also achieve a higher mean reward than repeated planning with the same reconstructed models, perform comparably to methods trained with controlled interactions, and generalize to systems absent from policy training after embedding inference alone. Applications to neural-behavioral recordings demonstrate suppression of predicted movement under constrained neural perturbations. Together, these findings show how shared dynamical representations support transferable control and mechanistic hypothesis generation from action-free recordings.

Fast control of long-lived quantum registers often relies on an auxiliary degree of freedom, a mediator, but real mediator excitation can introduce decoherence, leakage, or loss when the mediator is not the coherence-optimized component. We introduce dispersive interaction via analytical linear inversion (DIAL), an analytical control construction for excitation-suppressed interaction engineering in mediator-coupled quantum registers. In the dispersive regime, we show that off-resonant driving of an effective two-level mediator transition yields a diagonal Hamiltonian whose Pauli-\(Z\) interaction rates depend linearly on the drive intensities through a transfer matrix determined by the register-resolved transition spectrum. This establishes a direct linear map from experimentally accessible controls to induced multi-qubit interactions, enabling control design without iterative propagation of the full driven dynamics. We turn this structure into the target-aware DIAL algorithm, which selects favorable off-resonant tones and determines their drive intensities for a prescribed target interaction. We benchmark the resulting controls across 100 register realizations for each register size \(n=2,3,4,5\), targeting the highest-order interaction \(Z_1\cdots Z_n\) with an interaction phase of magnitude \(\pi/4\), which generates an \(n\)-qubit GHZ state up to local equivalence. Propagation under the full driven mediator-register dynamics yields typical gate fidelities above \(0.997\), while the maximum transient mediator excitation remains near \(1\%\) on average, far below what is typically possible with resonant Rabi drives. The manuscript is accompanied by an open-source Python package implementing the target-aware DIAL algorithm and the associated numerical validation workflow.

Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks. To address this challenge without sacrificing creativity, we propose HARPO, a reinforcement learning framework designed to jointly optimize faithfulness and creativity. HARPO incorporates a Hallucination-Aware Generative Reward Model (HA-GRM), trained via verifiable feedback, to assess both faithfulness and writing quality. A Selective Activation Mechanism (SAM) activates writing rewards only for outputs judged hallucination-free by HA-GRM, while a data curriculum progressively shifts training from creative writing to hallucination-centric tasks. On RAGTruth, our Qwen3-4B-based HA-GRM achieves a response-level F1 score of 78.08%, compared with 66.37% for the supervised fine-tuning baseline. Experiments on Qwen2.5 and Qwen3 models from 1.7B to 8B parameters show improvements in both faithful generation and writing quality. On Qwen3-4B, HARPO reduces the HA-GRM-judged hallucination rate on MultiHopRAG from 3.29% to 1.02%, while increasing the Arena-Hard-v2.0 creative-writing score from 16.95% to 27.54%.

We establish comparison Pogorelov estimates for admissible solutions of graphical $\sigma_k$-curvature equations, extending the two-surface estimate of Qiu and Yan for the graphical scalar curvature equation. The comparison graph is assumed only to be $k$-admissible, with a bounded slope and a positive interior gap that vanishes on the boundary. The estimates cover $2\le k<n$ for prescribed data $f(x,u)$, and $n/2\le k<n$ for general data $f(x,u,Du)$. The proof combines the two-surface localization and mixed Gårding comparison of Qiu and Yan with Yan's spectral concavity inequality and an angle-function weight.

Multi-chip-module (MCM) GPUs offer a promising path to scale compute capability beyond monolithic designs by integrating multiple chiplets on a common package. However, the impact of disaggregation on performance scalability and energy consumption remains underexplored. The design space grows rapidly across dimensions such as SMs per chiplet, chiplet count, and interconnection network. The inter-chiplet network is particularly critical, as it determines whether additional compute resources translate into performance gains. This limited understanding leaves industry and research without clear guidance on the performance and energy trade-offs of MCM GPU scaling. In this work, we investigate whether distributing compute and memory capability across multiple chiplets offers a more scalable alternative to concentrating resources. We quantify their effects on performance, energy, and efficiency and examine how inter-chiplet topology influences scalability at different system sizes. Our results demonstrate that a 16 chiplet Torus configuration with 256 SMs delivers a remarkable $2.40\times$ performance improvement over a state-of-the-art MCM architecture with the same compute capability, while simultaneously reducing energy consumption by $4.45\times$. These substantial gains provide evidence that disaggregation is a first-order architectural factor and will be critical to unlocking the performance and energy-efficiency potential of next-generation GPUs.

In Rational Extended Thermodynamics, the monatomic limit of polyatomic gases is often singular. For the internal-state density $\phi(I)=I^a$, with $a=(D-5)/2$, we prove that the normalized equilibrium distribution of the molecular internal energy $I$ concentrates at $I=0$ as $D\to3$, providing a rigorous justification of the Dirac prescription and its first-order correction. In classical ET14, the internal-mode energy differs from equilibrium by $-3\Pi/2$; positivity yields $-p<\Pi<(D-3)p/3$, while the exact entropy develops a boundary corner at $\Pi=0$. Negative dynamic pressures remain admissible, but fixed negative values do not approach the monatomic state with the prescribed total energy. This explains the need for compatible initial data. In the relativistic case, the internal energy is conditionally gamma distributed at fixed momentum; the constitutive integrals follow directly, and we derive an exact energy identity and the first correction to the Synge energy.

The rapid integration of renewable energy sources is transforming modern electric power systems into converter-dominated networks characterized by complex multi-timescale dynamics and new stability interactions over a wide frequency range. Accurate analysis of these phenomena requires high-fidelity models of transmission networks and power electronic converters, which often result in large-scale differential-algebraic systems with significant computational burden. This paper provides an analytical review of model order reduction techniques for converter-rich power systems, with particular emphasis on the key equations employed in the reduction process. Advantages and limitations of commonly used strategies, including timescale-based singular perturbation, projection-based, operator-based, data-driven, and machine-learning based methods, are discussed. The synthesis highlights key challenges related to harmonic interactions and multi-frequency dynamics, and outlines directions for the development of computationally efficient yet dynamically accurate reduced-order models for future power systems.

Knowledge-driven neural vocoders struggle to learn reliable fundamental frequency end-to-end, because spectral objectives provide weak supervision of periodic structure and lack phase information. We address this with a source-filter model whose alias-free additive source makes the instantaneous phase of the glottal cycle explicit; differentiating it yields the instantaneous frequency, and thus $F_0$, without an external tracker. Waveform error supervises only the deterministic harmonic path, while a spectral loss covers the full signal. On M4Singer and LM-SSD, the reconstruction is phase-aligned, reaching a signal-to-reconstruction-error ratio of 8.1 dB, while neural baselines remain negative. However, GOLF, given an external $F_0$, still reaches lower spectral distortion. On LM-SSD, the recovered $F_0$ attains the highest overall accuracy of any method tested, including supervised neural pitch trackers applied off the shelf, and the glottal closure instants come within 0.53 points of REAPER's identification rate, without any $F_0$ label.

Relational foundation models are increasingly pretrained on synthetic databases, yet downstream benchmarks reveal little about why one synthetic corpus produces a better model than another. In particular, strong performance may arise from realistic row-level statistics without the model ever learning to use relational structure. We study this as a data-attribution problem: which property of synthetic pretraining data induces relational computation? Using four Relational Transformer checkpoints trained with the same architecture, initialization, objective, and compute budget on corpora produced by four relational data generators, we trace a measurable property of the data to learned computation and downstream behavior. We hypothesize that relational mechanisms emerge when cross-table information is predictively necessary for the masked-cell pretraining objective. RelDiff exhibits by far the largest predictive gain from foreign-key-linked parents, and its corresponding model is uniquely sensitive to foreign-key interventions on unseen databases. This dependence survives a random-initialization control, grows monotonically with the fraction of corrupted links, and localizes to a serial cross-table pathway. Finally, disrupting the same mechanism during downstream inference removes RelDiff's advantage on relational tasks while leaving structure-insensitive models nearly unchanged. These results connect a property of synthetic training data to a learned mechanism and, through intervention, to downstream behavior.

Large language models are increasingly used as reasoning components in AI-driven materials Co-Scientists, yet the reliability of the resulting verification pipeline remains unclear. Metal-organic frameworks (MOFs) provide a particularly challenging setting because structures may appear under different identifiers, synthesis outcomes depend strongly on experimental conditions, evidence is distributed across heterogeneous sources, and some hypotheses require computation rather than literature alone. We introduce a diagnostic benchmark with four task families covering structural grounding, synthesis-condition verification, evidence-sufficiency verification, and MLIP-based computational verification. T-MOF-1-3 are evaluated under closed-book, retrieval-enabled, and oracle-evidence settings to localize failures in knowledge access, evidence acquisition, and reasoning, while T-MOF-4 separately evaluates computational verification. Guided by these diagnosed failure modes, we develop MOF-Verify, a failure-aware agentic harness that targets structural, literature, evidence-sufficiency, and computational bottlenecks before producing a final verdict. Across multiple backbone LLMs, MOF-Verify improves the reported average over direct inference and retrieval-based baselines on this benchmark, with gains concentrated in structural grounding and evidence sufficiency. Benchmark records and release-status documentation are available at https://github.com/IMMS-Ewha/MOF-Verify-Benchmark.

A coding agent can earn a passing grade by fixing its code, or by deleting the test that exposes the bug. Detecting such reward hacking requires recognizing attempted shortcuts, including those that fail. We release 173,561 annotated multi-turn coding trajectories from Qwen3-8B and show that supervising shortcut behavior independently of exploit success substantially improves detection. We introduce HACKTRACE, a behavior-supervised monitor that reads the internal states the agent already computes while generating code. Reusing these states enables monitoring before a turn is complete, without additional language-model tokens or passes. Combining this evidence with static features of the final files achieves a mean per-problem AUC of 0.997 with 8 ms of monitoring overhead, improving both accuracy and latency over monitors that run the model again on an honesty question and answer. The same generation states also provide an inexpensive monitoring signal for reinforcement learning. With strong GRPO penalties, HACKTRACE reduces the cheating share of passing solutions from 82-91% to 1-5%, while retaining honest, correct solutions and maintaining high detection accuracy as the policy evolves. Our results show that both the supervision target and the source of monitoring evidence matter for turning accurate detection into a useful training signal.

Clustered Federated Learning (FL) partitions a client population into groups of similar local distributions and trains one specialized model per cluster, mitigating client drift that degrades single-model methods under non-IID data. Prior methods discover cluster structure inside the training loop through gradient similarity, loss evaluation, or EM-style updates, thus increasing communication overhead, exposing gradients to inversion attacks, and providing no mechanism to assign clients absent from training. We propose RIPPLE, a clustered FL framework in which cluster assignment is computed entirely offline from a spectral characterization of each client's local data: a variance-weighted principal-component prototype embedded via the Wavelet Scattering Transform and decoded by a Gaussian Mixture VAE trained server-side on synthetic client populations before federation begins. Per-round communication cost matches FedAvg exactly, and a client absent from training obtains a personalized model from a single forward pass, without gradient computation, model evaluation, or extra communication round. We prove that the gap between RIPPLE's surrogate clustered objective and the oracle is bounded by a computable quantity decaying with client sample size and independent of federation duration; per-cluster convergence matches the minimax-optimal rate for non-convex smooth objectives. Across five benchmarks spanning controlled and realistic heterogeneity, RIPPLE consistently outperforms all baselines, with margins growing on the most realistic partitions.

The human menstrual cycle is regulated by complex hormonal feedback mechanisms which are essential for reproductive health. Existing mathematical models either rely on simplified phenomenological assumptions or achieve physiological accuracy at the cost of high mathematical complexity and computational costs. This paper presents a novel mechanistic model which balances biological realism with mathematical simplicity. The proposed framework describes the dynamics of the four central cycle-driving hormones, namely follicle stimulating hormone, luteinizing hormone, estradiol, and progesterone, using a system of ordinary differential equations. Mechanistic detail is incorporated through representations of feedback mechanisms, follicular development, luteinization, vascularization of the corpus luteum, and aspects of hormone synthesis mechanisms. Despite its comparatively low complexity, the model reproduces characteristic trends of hormone activity and concentration dynamics reported in experimental data. Due to its computational efficiency and modularity, the model is well suited as a basis for extended in-silico experiments, including applications in pharmacology and personalized medicine. Our results demonstrate that targeted incorporation of biological mechanisms enables accurate yet tractable modeling of endocrine system dynamics.

Large language models (LLMs) contain broad knowledge, but they cannot access all of it reliably. We study this problem through knowledge accessibility, which describes whether the knowledge needed for a query can be recalled from the model. We find that knowledge accessibility has a simple geometric structure in the model's representation of the query alone, before any generation. More accessible queries are closer to a center in the representation space, while less accessible queries are farther away. This geometry reveals a knowledge boundary that separates more accessible queries from less accessible ones. Accessibility consistently decreases with distance from the center, and this distance-based ordering transfers across datasets even when the centers differ. Controlled experiments further show that the centered geometry is more closely related to knowledge accessibility than to reasoning difficulty. The geometry also reveals when different interventions are useful. Query rewriting helps more for accessible queries, chain-of-thought reasoning helps more near the boundary, and retrieval gives larger gains beyond the boundary. These findings not only provide a new geometric view of how knowledge is organized in language models, but also suggest a useful pre-generation signal for adaptive inference.

Social bees are important pollinators that support biodiversity and crop pollination globally and serve as important model systems for collective behavior, but scalable measurement of individual- and colony-level behavior remains difficult in dense, occluded nest environments. Existing monitoring workflows use fiducial tags (e.g., ArUco) to preserve individual identity, yet tag-based tracking can fail when markers are obscured and provide limited information about body extent, spatial context, and untagged individuals. We present BeeWhere, an AI-assisted annotation and analysis workflow that combines ArUco detections with deep-learnt instance segmentations to quantify bumble bee behavior from high-resolution colony images and videos. Using bumble bee (Bombus impatiens) microcolonies as a test case, we annotate 483 frames containing 8,443 bee instances. We additionally annotate pollen balls, nest structures, and chamber boundaries, and train YOLO instance segmentation models for downstream behavioral analysis. Instance segmentations enable quantification of important behavioral metrics based on body contours, including nearest-neighbor distance, proximity to nest structures, spatial occupancy within the nest, and detection counts over time. We apply the BeeWhere models to tag-based tracking in an exploratory validation study assessing the behavioral impacts of neonicotinoid pesticide exposure. BeeWhere increased detection rates compared to tag-based tracking, particularly when bees were partially obscured or under challenging imaging conditions, and also captured treatment-associated changes in bee spatial organization not captured using tag-based tracking alone. These results suggest that instance segmentation can complement fiducial-marker tracking by recovering behaviorally meaningful signals under challenging colony conditions.

We introduce the skew Schäffer type constant as an infimal counterpart of the recently developed skew James type constant \cite{RaoWuLiuNiWang} and establish its basic geometric theory in Banach spaces. The new constant admits sharp universal bounds and characterizes uniform nonsquareness through a strict gap condition. Exact values are obtained in Hilbert and Lebesgue spaces, where an exact relation with the skew James type constant is established. We further derive a James-modulus estimate leading to sufficient conditions for uniform normal structure.

Nonlinear metasurfaces (NMSs) enable simultaneous generation and manipulation of harmonic light with ultrathin nanophotonic platforms. However, conventional NMSs typically adopt a monolithic single-layer configuration in which frequency conversion, polarization selection and wavefront engineering are intertwined to simultaneously occur within the same nonlinear meta-atoms, making these functionalities intrinsically interdependent and imposing constraints on material choice, structural symmetry, and the available geometric parameter space. Here, we propose and experimentally demonstrate a metasurface-on-lithium-niobate architecture that physically separates harmonic light generation and wavefront manipulation. In this platform, a z-cut thin-film LN layer serves as a spin-selective second-harmonic source through its intrinsic bulk second-order nonlinearity, while an overlaid dielectric metasurface subsequently and independently molds wavefronts of generated harmonic fields. The two-step process, by virtue of harmonic generation being decoupled from wavefront shaping, removes symmetry constraints associated with conventional nonlinear meta-atoms and enables the direct transfer of well-established linear metasurface design strategies to harmonic wavefront engineering. As proof-of-concept demonstrations, we realize spin-decoupled second-harmonic beam steering and spin-multiplexed holography, achieving high-contrast routing of harmonic signals with orthogonal spin states into distinct diffraction channels and high-fidelity reconstruction of independent holographic images. By redirecting the harmonic wavefront control from the meta-atom engineering to the modular system-level design, our approach establishes a versatile route toward integrated nonlinear meta-optics and provides a scalable platform for advanced optical functionalities.

Let $L$ be a line bundle of negative degree on an elliptic curve $E$. While the ruled surface $M := \mathbb P(\mathcal O_E \oplus L)$ is one of the simplest varieties in algebraic geometry, its tangent bundle $T_M$ is surprisingly complex: we first show that the ring of symmetric tensors is not finitely generated. We then construct a holomorphic function on the canonical extension $Z_M$ which allows us to show that $Z_M$ is not Stein.

Despite never being supervised on explicit 3D motion, large-scale text-to-video diffusion models synthesize realistic human motion in their generated videos. We ask whether this implicit knowledge can be turned into explicit 3D motion generation, without training a separate motion model. Probing a frozen Wan2.1 reveals that a recoverable motion signal is present in its intermediate states across the entire denoising schedule, not confined to the clean output. Motivated by this, we introduce parasitic co-denoising, a paradigm in which motion is decoded from the host model along its denoising schedule rather than produced by an independent generator. We instantiate it as the Parasitic Motion Decoder (PMD), an efficient flow-matching decoder that shares the host's noise schedule and reads its intermediate features through a $\sigma$-adaptive multi-layer fusion, leaving the host unmodified. Drawing its coverage from the host rather than from motion data, PMD leads dedicated motion generators on text-motion alignment at a small fraction of their trainable parameters, while producing paired video and motion in a single pass that motion-only baselines cannot match.

Long-term permafrost warming on the Tibetan Plateau is continuously and broadly reshaping cold-region landscapes, and has produced numerous localized and clustered retrogressive thaw slumps (RTSs). However, the question why RTSs are highly localized and clustered under the widespread permafrost warming remains incompletely understood. To understand and explain the RTS localization and clustering, we propose the Hillslope Weakening Field (HWF), which represents the spatial distribution of relative hillslope weakening states jointly formed by multiple environmental conditions, including thermal conditions, moisture and soil properties, topographic setting, permafrost and ground-ice conditions, and land-surface cover. We examined the correspondence between the HWF and RTS spatial selectivity across seven study regions on the Tibetan Plateau. Positive HWF-RTS correspondence was observed in six of the seven regions, with consistently strong relationships in five. In these five regions, RTSs preferentially occurred on more strongly weakened hillslopes, with occurrence rates 3.35-5.09 times the regional background rate within the 10% most strongly weakened slope units. Moreover, high-weakening hillslopes were not spatially isolated but correlated and connected across the landscape. RTS-bearing slope units and their surrounding hillslopes showed higher weakening levels than RTS-unrecorded slope units and their surroundings, and the weakening contrast declined with distance. These results suggest that RTSs preferentially occur on more strongly weakened hillslopes and cluster where high-weakening states are spatially continuous and connected. Our findings show how spatial differences in hillslope weakening states can help explain why geomorphic responses to widespread permafrost warming are highly heterogeneous across permafrost landscapes.

Heterogeneous parallel error detection architecture has been widely studied for safeguarding OoO superscalar processors in safetycritical systems, as it achieves significantly lower hardware overhead compared to traditional LockStep, by exploiting the parallelism that exists in a secondary execution. However, previous works do not cover the protection of privileged-mode execution, impeding their effectiveness in real-world deployment. Moreover, naive extension to privileged-mode can cause a litany of issues, from abysmal performance due to high synchronization costs, to full deadlocks. Here, we present PEEK, the first privileged parallel error detection architecture. Based on a deep analysis of privileged execution, we redesign the verification pipeline, addressing all the bottlenecks and bugs identified in privileged-mode protection. Evaluated using various metrics on an RTL-level full system running Linux, PEEK achieves full-privilege protection on Linux with negligible performance slowdown and affordable hardware overhead. PEEK has been taped out using a 28nm process, and its source is available at https://anonymous.4open.science/r/PEEK-3000.

We present high-cadence, multi-frequency monitoring of the blazar OJ 287 using KVN (22–129 GHz), ALMA (91.5–343.5 GHz), and Metsähovi (37 GHz), covering 2012–2023, together with published optical polarimetry. Within this decade-long dataset, an exceptionally large and smooth EVPA rotation is observed only during 2015–2017. The millimeter-band EVPA rotates by more than $\sim$300 deg over $\sim$1.5 yr, with a comparably large and more rapidly evolving rotation observed at optical wavelengths, while remaining comparatively stable at other epochs. The rotation coincides with a sequence of strong radio flares from 37 to 343 GHz whose peak amplitudes increase toward 2017 March, when the rotation ends. Modeling the ALMA 91.5 GHz light curve yields variability Doppler factors of $\delta_{\rm var} \sim 8$–$11$ during the EVPA-rotation interval, with no systematic increase across the flare sequence. The absence of a monotonic change in $\delta_{\rm var}$ indicates that progressively enhanced relativistic beaming is unlikely to be the primary driver of the rising flare envelope. The confinement of the large EVPA rotation to this interval, together with the nearly constant $\delta_{\rm var}$, indicates that the 2015–2017 event occurred when a single newly ejected disturbance temporarily dominated the polarized emission while propagating through structured inner-jet regions threaded by a helical magnetic field under modest jet-direction changes. Large EVPA rotations therefore arise intermittently as the jet orientation evolves, and are realized only when emission dominance and viewing geometry align favorably.

Context. The existence of tidal binary clusters (TBCs) in the Milky Way has been repeatedly suggested using Gaia DR3 astrometry and recent catalogs of binary clusters and groups. However, the dynamical feasibility of forming truly bound binary clusters through tidal capture remains uncertain. Aims. We perform a systematic reassessment of twenty star clusters, currently claimed as members of TBCs from the catalog of binary and grouped clusters of Liu et al. (2025) to ascertain their classification. Methods. For each candidate pair, we use Gaia DR3 astrometry and photometry to perform a simultaneous and self-consistent opti mization of age, distance modulus, reddening, and metallicity. Results. Our analysis shows that all studied TBCs fail in the required physical condition that ages are compatible with independent formation. After applying strict evolutionary criteria, these systems are more naturally explained as co-eval members of primordial cluster pairs or groups rather than products of tidal capture. Conclusions. We conclude that genuine tidal binary clusters, if present in the Milky Way, must be extremely rare. This work un derscores the importance of robust age determinations and physically motivated selection criteria in future searches for dynamically assembled cluster pairs.

One of the promising applications of wireless sensor networks (WSNs) is monitoring of the human body for health concerns. For this purpose, a large number of small sensors are implanted in the human body. These sensors altogether provide a network of wireless sensors (WBANs) and monitor the vital signs and signals of the human body; these sensors will then send this information to the doctor. The most important application of the WBAN is the implementation of the monitoring network for patient safety in the hospital environment. In this case, supporting patients' mobility is one of the basic needs, which has been underestimated in recent studies. The problem that involves providing the required energy for the units used in this type of network is challenging; for this reason, sent/received units with very low power consumption and with a very small radius are used in order to save energy. The resulting small sending range leads to the lack of support for patients' mobility. In this paper, the ad hoc mode is suggested for use to establish a network and a multipath routing algorithm for the purpose of supporting patients' mobility in a hospital setting. The results of the simulation show that, in addition to supporting patients' mobility, the use of the proposed idea instead of previously presented protocols reduces delays in data transmission and energy consumption; it also increases the delivery rate depending on the destination and the lifetime of the network, while increasing routing overhead.

We study a forward-backward McKean-Vlasov system with jumps and one-way coupling: the forward equation is independent of the backward variables, whereas the BSDE depends on the forward state and its law. A source generated by a maximal monotone operator is evaluated along the forward state and then accumulated in the backward equation. The operator is the subdifferential of a convex potential evaluated at the state translated by a function of the control, plus a term depending on the law of the state, so the control shifts the discontinuity interface of its minimal-norm selection. We establish modelwise well-posedness and strong stability, with estimates uniform over the model family. We then study the Yosida approximation: strong convergence, Minty-type strong-weak closedness when the state, its law, and the control (hence the moving interface) vary, and uniform convergence under an integrability condition. For an asymmetric loss-aversion prototype, a quantitative occupation estimate for a layer around the interface yields an explicit Yosida convergence rate. When the continuous martingale part is nondegenerate and the interface moves absolutely continuously, the occupation time is linear in the layer width. The same rate controls the backward components and the cumulative source.

We present and theoretically analyze a method for direct estimation of the second-order correlation function $g^{(2)}(0)$ from the click statistics of a multiplexed detector composed of M binary on-off detection channels. The method uses linear estimators of the first and second moments of the photon-number operator and can provide accurate results even for optical fields with substantial mean photon numbers. For a fixed number of channels, the detection range can be extended by introducing additional losses before the detector. We demonstrate the performance of the method for coherent, thermal, amplitude-squeezed, and photon-added coherent states. We also derive a nonclassicality criterion based on the estimated correlation function and show how the method can be adapted to unbalanced detectors. Our work establishes direct estimation of $g^{(2)}(0)$ as a simple and versatile method for quantitatively characterizing quantum light using multiplexed click detectors.

Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LLMs), but they introduce high inference-time overhead, with latency dominated by sequential decoding. We propose HyperThink, a text-to-parameter approach that amortizes this reasoning computation into a single query-conditioned parameter update: a lightweight hypernetwork reads the question and predicts updates to a small subset of the base LLM's parameters, while a vector-quantized decoder constrains them to a finite set of reusable patterns to improve robustness and transfer. Trained end-to-end on outputs from the base model itself, HyperThink eliminates long thinking traces at test time: after one hypernetwork forward pass, the adapted model generates a concise step-by-step solution and final answer without an intermediate trace, using far fewer tokens while retaining strong reasoning performance. Empirically, HyperThink improves the low-latency region of the accuracy-latency trade-off on mathematical and general reasoning tasks, with its strongest gains in the near-non-thinking regime.

Let $a(n)$ be the number of standard Young tableaux of shape $(n,n,n)$ whose entry in row $2$, column $1$ is odd; equivalently, the number of those whose first descent is even. This is entry A011553 of the On-Line Encyclopedia of Integer Sequences, contributed in 1996, and after thirty years of curation it carries no formula. We supply one, $a(n) = 8\,\bigl(n!\,(n+2)!\bigr)^{-1}\sum_{m=1}^{\lfloor n/2\rfloor} m(m+1)(3n-2m-1)!/(n-2m)!$, and use it to settle both of the conjectures the entry records. A creative telescoping certificate shows that $a$ satisfies a linear recurrence of order two with polynomial coefficients; the order-three recurrence conjectured by R. J. Mathar in 2023 is a left multiple of it, with explicit cofactor $(4S^{-1}-3)/(7n-9)$. We also prove $a(n)\sim 3^{3n+7/2}/(64\pi n^{4})$, the asymptotic conjectured by V. Kotesovec in 2014. The second proof gives slightly more than the conjecture asks: the position of the first descent has a limiting distribution, the probability that the $(2,1)$ entry equals $r+1$ tending to $r(r+2)/3^{\,r+1}$. Summing the even terms, a uniformly random tableau of shape $(n,n,n)$ has an odd $(2,1)$ entry with probability tending to $27/64$, whereas for a uniformly random tableau of $n$ cells of unrestricted shape the corresponding limit is $1/e$.