Papers

Time series (TS) anomaly detection (AD) lacks a universal detector, with the best choice determined by the series rather than by the algorithm. Existing methods therefore either select one detector for each series or average a pool of them. However, both assume that no label ever arrives, while a deployed monitor does receive them, since an operator confirms the alarms it raises. In this paper, we propose ONESHOT, which uses a single confirmed anomaly to weight the pool rather than to choose one detector from it. Specifically, we score each detector on the confirmed anomaly, and those scores become the weights of a soft mixture over the whole pool. We show that a single confirmed anomaly is enough to come close to the best detector for the series. Across 152 univariate and 73 multivariate series, ONESHOT outperforms the strongest single detector, averaging the whole pool, and even picking the best detector on the confirmed event. Code is available at: https://github.com/seunghan96/one_shot.

We propose to measure hard exclusive $\alpha$-knockout reactions from the light $\alpha$-conjugate nuclei $^{12}$C, $^{16}$O, and $^{20}$Ne using electron scattering. Three reaction channels will be measured for the first time on the same target set: (i) the quasi-elastic knockout $(e, e' \alpha)$ knockout, which establishes a baseline against $\alpha$-clustering model calculations and tests final-state-interaction (FSI) models; (ii) coherent deeply virtual Compton scattering (DVCS) on a bound $\alpha$, which provides a quark-level probe of the bound system through the beam-spin asymmetry; and (iii) coherent vector-meson production ($\phi$ and J/{$\psi$}) on a bound $\alpha$, which probes the gluon distribution of the bound cluster through two-gluon exchange. The central observables are ratios of cross sections and beam-spin asymmetries between the bound $\alpha$ in $^{12}$C, $^{16}$O, and $^{20}$ Ne and the free $^4$He ground state. Together these measurements directly address the question of the role of QCD in the structure of light nuclei : do the quark and gluon distributions of an $\alpha$ embedded in a heavier nucleus differ from those of a free 4He, and do those modifications track the increasing $\alpha$-clustering strength along the $\alpha$-conjugate chain? The measurement will use the 12 GeV CEBAF accelerator in Hall B with the CLAS12 spectrometer instrumented with the ALERT low-energy recoil detector and high-pressure gas straw targets. ALERT, which replaced the CLAS12 silicon vertex tracker, was successfully commissioned and operated during Run Group L (RG-L) from April through August 2025. It uniquely provides identification of low-momentum recoiling light ions (p, d, t, $^3$He, and $^4$He) close to the beamline at the high luminosities required by these exclusive channels.

Temporal audio grounding requires locating every occurrence of a queried target, making both occurrence counts and interval boundaries essential. We propose RefTAG, a post-training framework for verifiable temporal self-correction. Its Reflection Chain-of-Thought (Re-CoT) protocol combines draft intervals, discrete count and boundary actions, and revised intervals in one response. Supervised fine-tuning on these trajectories (Re-CoT-SFT) teaches the protocol; reinforcement learning (RL) jointly rewards final localization and revision outcomes. Across 8,228 queries, Re-CoT-SFT improves query-weighted mIoU over final-answer SFT by 1.74 percentage points. RL further improves TAG-Bench multi-target mIoU by 0.40 points and R@0.5 by 0.86 points. The final model achieves 38.50% TAG-Bench mIoU and 51.01% mean mIoU across three datasets, outperforming the evaluated open-source MOSS-Audio and FireRedAudio models and the commercial Gemini-3.1-Pro on both measures. These results demonstrate that explicitly coupling count- and boundary-aware revision with joint reward optimization provides an effective and verifiable path to temporal audio grounding.

We study the limiting distribution of the largest real part of the eigenvalues of truncated unitary matrices. Let \(U_m\) be an \(m\times m\) Haar-distributed unitary matrix, and let \(Z_1,\ldots,Z_n\) be the eigenvalues of its upper-left \(n\times n\) principal submatrix. We determine the limiting distribution of \(\max_{1\le i\le n}\Re Z_i\) under a unified normalization: the limiting law is Gumbel when $q:=m-n\to\infty$ and is of Weibull type when $q$ is fixed. The fixed-\(q\) limiting laws converge to the Gumbel law as \(q\to\infty\). We also give a unified formulation of the normalizing constants for the spectral radius that recovers the previously known limiting laws.

Genuine multipartite entanglement (GME) is a valuable resource for quantum technologies, but quantifying it in many-body systems can be demanding since measures defined through optimizations over bipartitions are often assumed to require large joint measurements and many measurement settings. We introduce the concepts of GME evaluation length and GME evaluation number, which capture, respectively, the largest subsystem that must be jointly measured and the number of distinct subsystems that need to be measured to evaluate multipartite entanglement measures defined via bipartition optimizations. Taking the genuine multipartite concurrence (GMC) as a canonical example, we show that, for $N$-qubit Haar-random pure states, the GMC can be evaluated using only $N$ single-qubit purity measurements. Thus, for typical many-body states, evaluating the GMC requires considering only the $N$ single-site bipartitions, rather than the exponentially many distinct bipartitions. We further corroborate this behavior numerically for physically accessible states generated by local random quantum circuits. Our results show that the practical cost of quantifying GME can be dramatically lower than suggested by its formal definition, establishing measurement complexity as a complementary operational lens for multipartite entanglement.

We determine the representation type and Schurian finiteness of the Tits algebra of a finite central real hyperplane arrangement over an algebraically closed field. The algebra is representation-finite when the arrangement has at most two hyperplanes, tame when it has three, and wild when it has at least four. Moreover, it is Schurian-finite if and only if the arrangement has at most two hyperplanes. Both classifications hold over algebraically closed fields of arbitrary characteristic. The proof combines Saliola's bound-quiver presentation with explicit module constructions, convex subquivers, essentialization, and deletion. In the $\mathbb{R}^2$ case, we describe all eleven indecomposable modules for two lines and construct a full exact embedding of modules over a free algebra for at least four lines.

Embedded platforms increasingly host safety-related streaming functions next to less critical ones. Whether a safety architecture pattern, such as a doer/checker pair, protects a function depends on the cores that run its components. Their schedule must meet the throughput requirement, and mixing safety integrity levels (SILs) on one core raises development cost. This report presents SafeDSE, a design space exploration framework for synchronous dataflow applications. One declarative constraint model decides the pattern of each actor, the placement that the fault model requires, the static-order schedule, and the platform, and prices SIL promotion. Guards activate the chosen patterns in a fixed graph that holds all of them, and node potentials encode throughput. An independent verifier re-checks every solution, and a generator turns its checks into evidence of a safety argument in Goal Structuring Notation. On 9 benchmark instances under period bounds, no rule that fixes patterns before mapping was both always feasible and never more expensive under conservative pricing, while joint exploration was. In the generated arguments, 49 to 62% of the leaves cite verifier checks; the rest are claims a mapping cannot discharge. Optimality within the model was proven for all synthetic graphs with up to 16 actors. This extended version adds the full model, the verifier and generator in detail, the complete catalog and parameters, and a proof of the potential characterization. It also reports every experiment per instance, a mutation analysis of the verifier, and the negative results of the development.

Spatial transcriptomics jointly captures molecular state, tissue morphology, and spatial organization, yet spatial-domain methods are typically compared only at late training checkpoints, obscuring differences in convergence efficiency. We introduce MP-MNCA, a morphology-prior masked neighbor cross-attention framework that preserves the full 3,000 dimensional highly-variable-gene representation while selectively routing molecular information across spatial neighbors, with histological similarity incorporated as a continuous attention prior under a self-supervised masked neighbor-contrastive objective. Across 12 heterogeneous human dorsolateral prefrontal cortex (DLPFC) sections, MP-MNCA is evaluated against established spatial-domain baselines using both convergence trajectories and epoch-wise spatial-domain visualizations. Our results show that MP-MNCA (i) reaches coherent spatial-domain structure in substantially fewer training epochs, as reflected quantitatively by convergence curves and qualitatively by progressive spatial grids, and (ii) maintains strong clustering quality, achieving a mean peak refined ARI of 0.546 versus 0.549 for the strongest baseline (STAIG), while attaining the highest mean refined NMI of 0.657 versus 0.640. Moreover, MP-MNCA outperforms STAIG on 7 of 12 sections in ARI and 11 of 12 in NMI, demonstrating faster convergence without sacrificing spatial-domain quality.

We study two related questions for nonlocal equations of Bellman type: the regularity of solutions when the set of controls depends on the state variable, and the periodic homogenization of stationary nonlocal Hamilton-Jacobi equations driven by uniformly elliptic integro-differential operators of order $\sigma\in(0,2)$, whose kernels depend simultaneously on the slow variable, the fast periodic variable and the control. The effective equation is again a nonlocal Bellman equation, but its supremum runs over a set of viscosity Mather measures that varies with the space variable, in a merely Hölder continuous way. We first develop a $C^{\sigma+\alpha}$ regularity theory for nonlocal Bellman equations with Hölder continuous, state-dependent index sets, extending the Schauder estimates of Caffarelli-Silvestre and Serra, in a general framework independent of homogenization. We then use these estimates as a source of compactness to construct a classical solution of the effective problem, unique among bounded classical solutions. Finally, combining this regularity with discounted approximate correctors, we make the perturbed test function method quantitative, and prove that the whole family of solutions converges to the effective solution, with the explicit rate $\epsilon^{\sigma\alpha/(2+\alpha)}$, where $\alpha$ is the Hölder exponent of the data.

A one-dimensional waveguide is a crucial medium for transferring propagating quantum information between remote qubits, yet the photon-mediated interaction is typically fixed by the static device geometry, qubit parameters, and their couplings to the transmission line. Such an interaction can be engineered to be coherent or dissipative via virtual- or real-photon exchange, respectively, yet its effective strength remains fixed in either case. Here we demonstrate dynamical control over this interaction by applying Floquet engineering to two superconducting qubits coupled to a waveguide terminated by a capacitive mirror. By positioning one qubit at an antinode and the other qubit at a node or an antinode of the resonant standing wave, we realize a Hermitian or non-Hermitian coupling between the two qubits. A periodic flux drive of the qubit frequencies then enables on-demand control of the coherent energy splitting and the collective decay. Our results establish Floquet engineering as a route to dynamically reshape both coherent and dissipative photon-mediated interactions between remote quantum emitters.

The Tarski fixed-point problem asks for a fixed point of a monotone map on a $k$-dimensional grid lattice of width $n$ and has applications in game theory, economics, and verification. Substantial gaps remain between the known upper and lower bounds on its quantum query complexity. We prove two hardness bootstrapping theorems for quantum query lower bounds. Starting from a hard instance family whose instances have pair-wise incomparable unique fixed points, and a lower bound established by the nonnegative adversary method, we bootstrap this lower bound by (i) a factor of $\Omega(t)$ when the dimension is multiplied by $t$ and the width remains unchanged, and (ii) a factor of $\Omega(\log m)$ when the width is multiplied by $m$ and the dimension is doubled. Applying these theorems, we obtain three consequences: (i) an $\Omega(k\log^2 n)$ lower bound for $k\geq 4$, improving previously best-known classical and quantum lower bounds; (ii) for every fixed constant $\alpha>0$, there exists a constant $\delta(\alpha)>0$ such that $Q_\epsilon(\text{Tarski}(n,k))=\Omega(kn^{\delta(\alpha)})$ whenever $k>n^\alpha$, and in particular, an $\Omega(kn^{0.0072})$ lower bound for $k>n^{2/203}$, the first lower bound with simultaneous linear dependence on the dimension and polynomial dependence on the width; and (iii) a superpolynomial lower bound of $(k/\log k)^{\Omega(\log k)}$ for exponentially wide grids, where $n=2^{\Theta(k)}$. The last result rules out a route to polynomial-time algorithms for simple stochastic games and ARRIVAL through generic polynomial-query black-box Tarski solvers.

Let $P$ be a short-range projector on $\ell^2(\mathbb{Z}^{2n},\mathbb{C}^m)$ and $U=(U_1,\dots,U_{2n})$ be a transverse family of open subsets of $\mathbb{R}^{2n}$. We give an elementary proof of a recent result of Martinian-Stoiber, which extends a two-dimensional result of Drouot-Zhu to all even dimensions. Specifically, using only Stokes' theorem and the Poincaré lemma, we show that the geometric Kubo formula $\operatorname{Tr}\big(P[P,\mathbf{1}_{U_1}]\wedge\cdots\wedge[P,\mathbf{1}_{U_{2n}}]\big)$ is the product of a topological integer $\mathcal{X}_U$, which depends only on the relative positions of $U_1, \dots, U_{2n}$, and of the bulk conductance of $P$. Our proof writes the geometric bulk conductance as the pairing of $P$ with a cochain on $\mathbb{Z}^{2n}$ - an idea originally due to Kitaev - and deforms the cochain rather than the sets $(U_1, \dots, U_{2n})$. The proof presented in this paper is due to Claude Opus 5.5.

Deterministic preparation of bosonic resource states is a central task in continuous-variable quantum information processing and quantum metrology. Here we develop a measurement-based imaginary time evolution (MITE) protocol for preparing squeezed and non-Gaussian states. The protocol combines ancilla-mediated weak measurements of the Hamiltonian with a conditional phase rotation that suppresses excited state trapping and stabilizes convergence toward the ground state. The method is potentially applicable in hybrid discrete variable-continuous variable (DV-CV) platforms, where the qubit acts as the measurement readout. We demonstrate deterministic convergence over a broad range of parameters and show that the performance degradation near the stability boundary is governed by the shrinking Bogoliubov quasiparticle gap. The same framework also extends to non-Gaussian bosonic resource states, including Schrödinger cat states and excited states of the single-mode squeezing Hamiltonian, which display Wigner negativity. The squeezed states remain robust under bosonic loss over experimentally relevant regimes.

Language plays a crucial role in research and technology design, throughout problem formulation, method descriptions, analysis of findings, and interpretation. In this paper, we focus on a particular, trending linguistic pattern in the Human-Computer Interaction literature, which we call -abilities. Examples include usability, accessibility, and explainability. These -abilities have become increasingly popular, with potential impacts on the discipline. They take a verb, turn it into an adjective, and finally nominalize it. We approach the -abilities phenomenon with a survey of the HCI literature between 1988 and 2025. All in all, we identified 349 different -abilities. Moreover, the use of -abilities notably increased since the late 2000s, both in the absolute number of different -abilities and the frequency of use of the most popular ones. The majority of -abilities we found had a positive notion, and most -abilities with a negative notion are prefixed counterparts of their positive opposite (e.g., un+sustainability). We critically discuss implications of their increased use. First, -abilities change the literal meaning of design intentions. For example, acceptability reduces the concrete verb accepting into only its potential (acceptable), and finally nominalizes it, claiming the existence of acceptability as a somewhat stable and real, abstract characteristic of a technology. Second, the prefix-based opposite construction can weaken the design goal. Un+sustainability avoids clarification of the -ability's meaning (here e.g., ecological or financial), and does not spell out the expected impact of designs that are not sustainable (e.g., the destruction of nature). Future work should compare the use of -abilities in English with other languages to better understand the ways in which they transform meaning.

We show that the Stein degree of horizontal divisors on generalised log Calabi-Yau fibrations or certain vertical divisors of Fano type fibrations over arbitrary fields of characteristic zero is bounded by constants that are independent on the ground field.

Autoencoders are the core component of latent video generative models, compressing high-dimensional videos into compact latent spaces for downstream generation. However, existing video autoencoders are primarily optimized for pixel-level reconstruction, providing limited supervision for learning structured latent representations. We argue that videos should not be treated merely as signals to be reconstructed, but rather as 2D projections of dynamic 3D worlds that exhibit explicit and well-structured geometric properties. Building on this insight, we propose Video Geometry Grounded Autoencoder (VGGA), a geometry-aware video autoencoder that explicitly incorporates geometric constraints into the reconstruction process. Specifically, VGGA grounds the learned representations along three complementary geometric dimensions: Structural Depth, Motion Correspondence, and Projective Warping. These objectives are unified within a Video Geometry Decoder, which employs a shared spatio-temporal upsampling trunk to jointly optimize geometric learning and RGB reconstruction. Through these geometry-grounded objectives, VGGA encourages the latent representations to capture the underlying 3D structure and motion of dynamic scenes. VGGA consistently improves video generation across both geometry-centric and human-motion domains, reducing gFVD by 28.1% on RealEstate10K and 37.7% on Taichi-HD. Notably, these substantial generation improvements are achieved while maintaining comparable reconstruction quality, demonstrating that geometry-grounded representation learning yields a more effective latent space for video generation.

We resolve a conjecture of Łuczak, Mieczkowska and Šileikis on the maximal tail probability of sums of independent and identically distributed nonnegative random variables. Specifically, for every integer $k\ge1$ and every $0\le x\le1/k$, we prove \[ \sup_{\mathbb E X_1\le x}\mathbb{P}(X_1+\cdots+X_k\ge1) =\max\{1-(1-x)^k,(kx)^k\}, \] where the supremum is over nonnegative i.i.d.\ random variables $X_1,\ldots,X_k$. We also characterize all maximizing distributions. Using this result, we prove the Erdős Matching Conjecture for every fixed $k\ge3$, all sufficiently large $s$, and every $n\ge k(s+1)$. More specifically, we show that in this range, the maximum number of edges in an $n$-vertex $k$-uniform hypergraph containing no $s+1$ pairwise disjoint edges is \[ \max\left\{ \binom{k(s+1)-1}{k}, \binom nk-\binom{n-s}{k} \right\}. \] Moreover, every extremal family consists either of all $k$-sets meeting a fixed $s$-set or of all $k$-sets contained in a fixed $(k(s+1)-1)$-set.

Spatially encoded speech conveys not only linguistic information but also spatial cues. This enables listeners to distinguish and localize speakers, contributing to more natural and realistic auditory experiences in applications such as remote conferencing and immersive media. Neural speech codecs (NSCs) enable high-quality reconstruction of monaural speech at increasingly low bitrates while preserving linguistic information, but their ability to effectively preserve spatial cues remains limited. We propose FOAdapter, a plug-and-play spatial tokenizer that leverages a pretrained monaural speech codec to efficiently represent both linguistic and spatial information in first-order ambisonic (FOA) speech at low bitrates. Comprehensive evaluation demonstrates that FOAdapter consistently preserves spatial information across all evaluated codecs. Source code and audio examples are available on our demo website.

Language models trained only on text from before a given year are useful for studying hindsight and historical language, but the open collections of pre-modern Chinese carry no publication dates and unreliable dynasty labels. We present Lüshi-1840, a 983M-parameter model trained from scratch on 12,262 books (1.28B characters) selected as composed before 1840. A first corpus, dated by a common heuristic, in effect ended at 1644 while containing twentieth-century material. We describe the re-dating and cleaning pipeline that replaced it and an audit with markers the pipeline did not use, which still finds later material. On the model side we measure surprisal, relative to a modern reference, on annals compiled after the cutoff. For events after 1840 against events up to 1820 the ratio is 1.38 (95% interval 1.33-1.44), against about 1.00 for modern general-purpose models. In a series of 294M models, adding the later court records lowers the ratio from 1.33 to 1.16, and the two models diverge from the 1820s, where the added records begin. The probe tracks the coverage of sources and does not establish a single cutoff year; none of our probes exposes the audited contamination. Model, book list and scripts will be released.

We study the intrinsic 0.5–2~keV emission of the hot circumgalactic medium (CGM) in the IllustrisTNG simulation TNG100-2 at redshifts 0, 0.1, and 0.2. We calculate the emission of gas cells with $T \geq 10^6$~K in the spherical shell between $0.15R_{500}$ and $R_{500}$ and compare samples selected by their star-formation activity. X-ray luminosity increases with halo mass, although the class-dependent fits cover different mass intervals. Among galaxies with positive luminosity and star formation rate (SFR), the high-activity sample retains a positive association after a linear adjustment for halo mass. The residual association is weaker in the quiescent subset with positive luminosity and SFR. Within $10^{11.75}<M_{500}/M_\odot<10^{12.25}$, the high-activity sample has more centrally concentrated emission and a median luminosity about five times that of the quiescent sample. The two populations have similar median hot-gas masses despite their different X-ray luminosities and radial profiles. These differences are consistent with variations in the spatial distribution and thermodynamic state of the emitting gas, although the samples are not matched in halo mass. Our results suggest that soft X-ray emission probes the structure of the hot CGM as well as its gas content, offering a way to investigate how star formation and feedback shape gaseous halos.

Freely swiveling casters are widely adopted in industrial AMRs/SDVs for their mechanical simplicity and load capacity. However, caster misalignment during low-speed or high-curvature maneuvers induces strong contact nonlinearities, widening the planning-execution gap between commanded $(v, \omega)$ and realized motion. Existing caster-aware strategies are highly sensitive to state-estimation reliability, causing unstable execution in uncertain transition regimes. To address this, we present Multi-Stage Caster-aware Control (MSC), a planning-execution integrated hierarchical framework featuring two core mechanisms. First, an uncertainty-aware NMPC (CU-MPC) dynamically schedules caster-aware costs and input regularization using predictive uncertainty to suppress overly aggressive maneuvers. Second, a reinforcement learning-based execution layer estimates and predicts caster states under partial observability, transforming high-level velocity commands into physically executable wheel speeds adapted to current contact conditions. Validations in Isaac Sim and real robots demonstrate that MSC offers a clear comparative advantage over existing decoupled compensation approaches by significantly improving caster-state estimation accuracy, minimizing misalignment, and enhancing tracking robustness in caster-sensitive maneuvers.

Quantum teleportation, a key building block for quantum technologies, enables distant parties, Alice and Bob, to transfer an unknown quantum state using entanglement, local operations, and classical communication. In this work, we propose a quantum teleportation protocol for single-photon time-frequency continuous variables, where time is the time of arrival at a detector. We detail the operational requirements of the protocol, notably the nonlinear interactions needed for the photon-photon time-frequency entangling gate. By explicitly treating the detector position as the reference frame for time of arrival, we show that, even with ideal classical and quantum channels, the input-output fidelity is not unity unless the receiving party can correct the detector-position mismatch. This results in a trade-off between teleportation fidelity and the temporal bandwidth of the input photon wavepacket. The protocol itself can be used for remote detector synchronization. In particular, we show that time-frequency cat and Gottesman-Kitaev-Preskill states provide an internal temporal ruler through fidelity revivals. We further derive the classical and best-copy fidelity thresholds for single-photon Gaussian wavepackets, finding that the relevant vacuum-equivalent shot noise is input-dependent, in contrast with quadrature-based continuous-variable quantum teleportation. Our proposal derives from a quantum theory of space-time events, whereafter we establish space as a reference for the time degree of freedom, thereby grounding our work in ongoing research on reference frames and superselection rules in quantum optics and quantum information. Beyond protocols harnessing time-frequency continuous variables of single photons, this work points toward remote synchronization in quantum communication architectures.

Uniform orbital detuning can restore spatial extent even as quasiperiodic modulation strengthens. In a trimer ladder with fixed hoppings and orbital modulation signs, increasing detuning and modulation amplitude together drives successive avoided crossings that redistribute orbital weights, suppressing the projected onsite potential in both lower molecular bands while leaving the third strongly modulated. Complete-band participation scaling supports two reentrant high-participation windows, with wave-packet propagation corroborating spatial recovery at representative controls over the studied sizes and times. These diagonal cancellations leave residual interband modulation, which generates a self-energy whose magnitude depends on the molecular level separation. Gaussian quadrature improves convergence of closed-walk determinant estimates used to cross-check energy-resolved attenuation. Comparing the coupled and decoupled ladders under the same finite-size participation criterion shows that restoring the full off-diagonal modulation narrows both sampled windows, with a larger absolute contraction of the second.

Automatic Speech Recognition (ASR) systems are commonly designed to transcribe linguistic content independently of speaker identity. However, speaker-specific information can persist within the latent representations of ASR models, raising privacy concerns when training data includes sensitive individuals. While machine unlearning has been studied in several speech tasks, applying speaker unlearning to ASR remains challenging due to the strong entanglement of linguistic and speaker-related information in speech representations. In this work, we conduct a layer-wise analysis of Self-Supervised Learning (SSL) based ASR representations, revealing that speaker identity is primarily encoded in intermediate encoder layers. Based on this observation, we explore a feature-level unlearning strategy that targets these identity structures. Experiments on the VCTK corpus show that the proposed approach removes the influence of specific speakers while maintaining robust transcription performance.

This paper studies near-field massive MIMO channel estimation based on an electromagnetic channel model with limited pilots. Starting from Maxwell's equations, we derive the homogeneous background response and the scattering response in the spatial and wavenumber domains. It is found that a compact scattering cluster produces a localized receive–transmit transverse-wavenumber block across finite apertures, thereby yields a block-sparse NLOS channel coefficient matrix in the wavenumber domain. Considering the spatial non-stationarity of large-scale antenna arrays, we introduce visibility factors at the transmit and receive sides of the channel model. To solve the non-stationary sparse channel estimation problem in the wavenumber domain, we propose a visibility-aware block-sparse Bayesian channel estimation algorithm. Simulation results demonstrate that the proposed method achieves lower channel-estimation NMSE than benchmark schemes.

One of the unsolved questions in astrophysics and cosmology is the development of a robust physical model to explain the thermal history of the intergalactic medium during the epoch of reionization. Various astrophysical models have been proposed, but current models remain only partially constrained by available observations. To clarify this ambiguity, hydrodynamical, semi-numerical, and N-body simulations can help solve this issue. One notable simulation from the epoch of reionization is the 21SSD simulations, which provide high-resolution results for the global 21-cm brightness temperature and the 21-cm power spectrum during the epoch of reionization. The simulation includes 45 models spanning three key astrophysical parameters of X-ray emissivity, Lyman-band emissivity, and the X-ray hard-to-soft emissivity ratio. Distinct values of these parameters result in different models of this simulation. In this work, four active radio observatories, including HERA, MWA, GMRT, and LOFAR have been considered to investigate the sensitivity and thermal noise of the 21-cm power spectrum. Each observatory has a particular sensitivity based on its instrumental characteristics, which helps assess which models lie above the forecast sensitivity thresholds of current instruments.

A compact metric tree is the natural domain of spatial data recorded along a river network, a root system, or the branched processes of a cell, and a suitable distance between two points is the length of the path joining them. That distance admits almost none of the covariance models of spatial statistics, since only the roughest members of each family stay nonnegative definite, and the resulting fields are nowhere mean square differentiable. The repair proposed here is a change of metric. Comparing two points through their distances to the leaves gives a Hausdorff-type metric built from the shortest-path one. The new metric has a closed form, agrees with the shortest-path one inside every edge, and is strictly smaller across branching points. Any correlation function valid in a Euclidean space of dimension equal to the number of leaves may then be composed with it, returning the Matérn, powered exponential, and generalized Cauchy families at every degree of smoothness, and, on trees with few leaves, also correlations that change sign or have compact support. The same construction yields the multivariate Matérn and space-time models of the Gneiting class. In simulations covering a range of smoothness and of spatial scale, maximum likelihood provides estimates without appreciable bias, and on a reconstructed microglial cell the smooth models reduce the cross-validated prediction error by half.

Variants of targeted maximum likelihood estimation (TMLE) and augmented inverse-probability weighting (AIPW) with machine-learning nuisance models differ in which observations train them, how bias is corrected, and whether splits are repeated. In known-truth simulations of a binary-outcome average treatment effect, eight headline estimators shared each setting's library, inner folds and truncation bound; coverage differences were judged against a prespecified two-percentage-point margin. With a low-complexity library and good overlap (n = 3000), all but a half-sample baseline covered equivalently (0.928-0.939). With one random forest on the primary data-generating mechanism, estimators fitting the propensity in-sample undercovered severely (0.12-0.48), whereas cross-fitting both nuisances raised coverage to 0.91-1.00, two overcovering. Under near-positivity, every AIPW estimator covered better than every TMLE estimator with the unweighted fluctuation (0.907-0.917 versus 0.656-0.806 at the primary setting, all undercovering); looser truncation widened the gap. At the primary settings, pooled-versus-fold-specific and single-versus-double cross-fit constructions (involving four additional estimators) were usually equivalent; under extreme weights, pooled-targeting estimates were 1.7-1.8 times as dispersed as fold-specific ones. Repetition reduced partition dependence but made several intervals more conservative. Two real-data analyses illustrate these sensitivities. Results support cross-fitting both nuisances when empirical-process conditions cannot be justified, and assessing truncation and AIPW-versus-TMLE sensitivity under poor overlap.

A model may select the correct answer yet fail to construct it through a sequence of local decisions. We introduce Choice2Seq, a diagnostic benchmark that measures correctness retention across complete-answer selection and sequential construction in independent runs on the same problems and candidate sets. Sequence Composition Rate (SCR) measures the fraction of correctly selected items that are also answered correctly through local construction. Jev achieves 97.1% and 99.5% selection accuracy on arithmetic and lexical retrieval, yet retains only 33.0% and 2.1% of those correct choices under serial construction, respectively. Substantial gaps also emerge for StartLux and Qwen, including when Qwen emits characters directly. Restricting local options to candidate-supported characters improves construction, while explicit position indexing improves copying fidelity, even for incorrect selections. These interventions distinguish task correctness, plan fidelity, and output compliance. Matched dialogue-state evaluations reveal that local field construction largely preserves Jev's correct states but reduces Qwen's mean-turn accuracy, highlighting task- and interface-dependent differences. Sequence-level confidence and request-cost analyses further characterize completed outputs.

As network communication increasingly becomes a performance bottleneck for ML and HPC workloads, it is more important than ever to optimize multi-GPU data transfers. Recently, low-level device APIs have emerged that allow programmers to perform fine-grain optimizations via one-sided put/get operations triggered directly from the device. However, the communication libraries providing these device APIs (e.g., rocSHMEM, MSCCL++, and NCCL/RCCL GIN) currently lack a rigorous memory model defining how their primitives interact with the underlying memory systems. Without this, emerging HPC and ML workloads aggressively optimizing network communication via device APIs can be brittle, error-prone, and face compatibility and inter-operability challenges. This paper proposes the use of API-level memory models for these emerging device communication APIs. An API-level memory model defines the memory semantics of communication API calls in a way that abstracts lower-level details from the API user and simplifies development. This work defines one such model for device-side communication (implemented in AMD's recently open sourced rocSHMEM library) and demonstrates how formal relaxations to such a model can enable low-level LLM performance optimizations while retaining a well-defined standardized interface. Specifically, we describe how updating the SHMEM API implemented by rocSHMEM can properly support cache bypassing for network data, improving performance of a collective communication workload by up to 2.48x, a fused compute+communication workload by up to 1.77x, and a state-of-the-art expert-parallel communication workload by up to 1.18x.