Papers

This paper proposes a movable-element (ME) enabled simultaneously transmitting and reflecting surface (ME-STARS)-assisted non-orthogonal multiple access (NOMA) communication system, where the positions of STARS elements can be flexibly reconfigured to introduce additional spatial degrees-of-freedom. Our objective is to maximize the sum achievable rate by jointly optimizing the active beamforming at the base station (BS) as well as the passive beamforming and MEs positions at the STARS. To tackle the resultant highly-coupled non-convex optimization problem, an alternating optimization-based algorithm is proposed, in which the original problem is decomposed into three subproblems. For the BS and ME-STARS beamforming subproblems, the successive convex approximation method is employed. For the MEs positions optimization, penalty terms are incorporated into the objective function to handle the constraints, thereby transforming the original problem into an unconstrained formulation. The gradient descent algorithm is subsequently applied to iteratively approach the sub-optimal solution. Numerical results demonstrate that: i) ME-STARS-assisted NOMA communications achieve superior performance compared to the NOMA communications assisted by STARS with fixed-position elements; and ii) the performance gain brought by the MEs is significant even with restricted moving region size.

We consider the recovery of the fractional order in a time-fractional evolution equation with a source of the form $p(t)f$, where the temporal factor $p$ is also unknown but belongs to a bounded admissible class and has a prescribed nonzero value at $t=0$. From one scalar observation we prove a uniform small-time expansion and obtain a Hölder estimate for the fractional order. More precisely, if a fixed derivative order $\gamma$ lies below every admissible fractional order, then the stability exponent is $\alpha_+/(2\alpha_+-\gamma)$; for the undifferentiated observation this gives the exponent $1/2$. We also show that differentiating the data to an order larger than all admissible fractional orders is supercritical: the corresponding derivative is unbounded near $t=0$ whenever the two fractional orders are different. A uniform estimate for the two-parameter Mittag–Leffler function, including the diagonal case, is proved in the appendix.

Real-time voice agents have moved from research prototypes to production deployments, yet the literature describing them is fragmented across three communities that rarely cite one another: speech foundation modelling, turn-taking psycholinguistics, and agentic evaluation. Architecture papers report latency, turn-taking papers report prediction accuracy, and agentic benchmarks report task success, so no single number describes whether a deployed agent is actually good. We address that gap with three evidence-based claims, each traceable to a corpus of 38 primary sources organised into an application-centric taxonomy of six categories. First, architecture choice is a deployment constraint rather than a settled verdict: a 2026 enterprise tutorial reports that no fully self-hostable end-to-end system yet meets production constraints, while a chunked cascade independently reaches state-of-the-art duplex behaviour, showing duplex behaviour is separable from duplex architecture. Second, evaluation has shifted decisively from component quality toward grounded outcomes, with recent benchmarks verifying backend state rather than trusting what the agent claims to have done. Third, the dyadic assumption in most models and benchmarks is breaking down: multiparty turn-taking and multi-speaker reasoning benchmarks show that deciding when not to speak, and reasoning about who may be told what, are first-class capabilities two-participant framings cannot measure. For each source we state the problem it targets, its mechanism, and its reported evidence, alongside the search strategy, inclusion criteria, and a verification step that caught a misattributed arXiv identifier in circulation. We propose TRG (Timing-Recovery-Grounded), a reporting standard characterising an agent by timing, post-disruption recovery, and state-verified outcome together, with a conditional fourth axis for multiparty deployments.

Text-based memory and context compression support reuse of past interactions. Resizing continuous memory changes the input to a frozen LLM, coupling capacity allocation with readout. We propose Hard-Origin Adaptively Softened Memory (HasMem). Frozen hard-prompt embeddings provide a verifiable initial state. A controller adjusts memory widths, a Writer re-encodes resized entries, and Reader and Global provide readout adaptation and cross-turn state. On all $535$ questions in a reconstruction probe derived from the Multi-Session Chat (MSC) development split, the main configuration achieves lexical F1 of $95.3$ ($+4.4$ percentage points) at $93.6\%$ of the hard reference's framed memory positions. With approximately matched per-question target body budgets, six configurations at mean per-entry retention around $0.83$–$0.91$ exceed rule-based re-encoding by $8.0$–$23.6$ exact-match (EM) percentage points. With fixed model parameters and rule target width ratio $0.75$, Global's EM gain passes a user-level exact paired test with Bonferroni correction over eight comparisons. On all $500$ LongMemEval-S questions, local lexical F1 rises from the hard reference's $3.4$ to $8.9$, and answer negative log-likelihood (NLL) falls from $12.257$ to $5.274$. F1 gains accompany lower EM on both evaluations.

Diagnostic consultation is an online sequential decision-making process in which clinicians gather evidence through patient interaction until a diagnosis is sufficiently supported. Automating this process requires adaptive inquiry and interpretable decisions. Bayesian networks offer a natural foundation by updating diagnostic posteriors as evidence accumulates, but their use in open-ended consultation raises two challenges: linking diagnostic hypotheses to potential inquiries and translating evolving posteriors into consultation decisions. We introduce AutoDisym, an automated pipeline that integrates diagnostic knowledge with heterogeneous diagnosis-labeled clinical narratives to construct a Disorder–Symptom Bayesian Network (DSBN). Building on the DSBN, we propose ConsultMind, an uncertainty-aware framework that updates disorder posteriors after each response and uses posterior uncertainty to guide inquiry and diagnosis. We evaluate both methods across psychiatry, respiratory medicine, fever clinics, and three public datasets. The results show that AutoDisym can automatically construct high-quality DSBNs and that ConsultMind consistently improves diagnostic performance and explanation soundness. For example, AutoDisym achieves macro-averaged F1 scores of 81.37 for canonical symptoms and 72.19 for manifestations using GPT-5.6-Sol. ConsultMind improves Top-1 and Top-3 diagnostic accuracy by up to 22.15 and 37.89 percentage points, respectively. Physician evaluation further shows that ConsultMind improves the quality of ranking explanations, differential diagnoses, and diagnosis rationales across LLMs of different scales. This work offers a promising approach to automatic diagnostic consultation.

Existing human motion diffusion methods provide strong motion generation quality, and recent style transfer models can inject target style cues, but fine-grained continuous control of style intensity remains underexplored. In production, style intensity is subjective across artists and directors, so the practical requirement is not a universal absolute unit, but a reliable monotonic control axis. We propose Motion Style Slider, a motion-to-motion style transfer framework for endpoint-supervised continuous control. Given a content motion and a style motion, we construct a style direction in a learned motion-style embedding space and condition diffusion generation with a scalar intensity. The training objective combines diffusion denoising with latent intensity regularization to encourage smooth and monotonic style scaling without requiring intermediate-intensity ground-truth motions. Our framework is compatible with pretrained motion diffusion backbones and supports heterogeneous style datasets, including the multi-actor style motion dataset. To test out-of-range usability, we additionally introduce a small real-capture over-reaction extension and evaluate large-intensity behavior against these unseen targets. Experiments measure controllability, interpolation/extrapolation behavior, content preservation, and motion realism, with ablations on direction construction and loss design.

We consider inverse problems for a time-fractional diffusion equation with a second-order elliptic operator and the homogeneous Dirichlet boundary condition in a bounded domain of $\mathbb R^N$. The initial value is unknown and is not assumed to be zero. The main result concerns the simultaneous determination of the fractional order and the time-dependent source factor from one weighted spatial observation. For irrational orders in $(0,1)$ and bounded source factors satisfying an exponential integrability condition, we prove that a nonzero observation given for all positive times uniquely determines both unknowns. We also describe the information about the initial value determined by this observation. A counterexample shows that uniqueness of the source factor may fail at the classical order $\alpha=1$, even when the observation is nonzero. In the second part, we establish two uniqueness results for the fractional order when the elliptic operators, initial values, and source terms in the compared equations may be different. The first result is proved by the Laplace transform, whereas the second uses short-time observations and the asymptotic behavior of the Mittag–Leffler functions.

We present a systematic spectral study of the X-ray emission from the Narrow- Line Seyfert 1 (NLS1) galaxy Ark 564 using Swift/XRT observations spanning sixteen years. The source exhibits significant variability in both the soft and hard X-ray bands, along with a strong positive correlation between them, indicating a closely coupled emission mechanism within the accretion flow-corona system. A simple absorbed power-law model is statistically insufficient to describe the spectra, confirming the presence of an additional soft component. To investigate the origin of the soft excess, we performed multi-model spectral fitting. Incor- porating a warm Comptonization component yields a statistically improved fit, supporting the scenario in which the soft excess arises from Comptonization in a warm, optically thick corona. We also applied a relativistic reflection model to representative observations; however, given the moderate spectral resolution and limited high-energy coverage of Swift/XRT, the current data do not allow us to robustly determine whether reflection provides a superior or physically preferred description. Our results indicate that warm Comptonization offers a consistent explanation for the observed soft excess in Ark 564 within the limitations of the available data, while deeper broadband observations are required to defini- tively distinguish between competing models and comprehensively characterise the long-term X-ray spectral variability of the source

We present a comprehensive photometric and spectroscopic analysis of the contact eclipsing binary BS Cas, based on 22-yr multi-site light curves and 2-yr high-resolution spectra. The long-term photometry shows pronounced year-to-year changes in the light-curve shape, including eclipse-depth reversals between the primary and secondary minima and variations in the O'Connell effect. The first double-lined radial-velocity (RV) curves of BS Cas are derived from the spectra using broadening-function (BF) profiles. Sharp peaks absent from the 2018 BFs appear in the 2020 BFs, suggesting an additional component along the same line of sight as BS Cas. Simultaneous modeling of the BVR light curves and the RVs shows that BS Cas is a contact binary in which the more massive component is cooler than its less massive companion, with a mass ratio of 2.515 and a fill-out factor of 32.3%. Multi-epoch modeling suggests that the long-term photometric variations arise from evolving starspots and changes in the third-light contribution. The eclipse-timing variation diagram confirms a secular orbital period decrease at a rate of $-2.305 \times 10^{-7}~\mathrm{days\,yr^{-1}}$ consistent with angular momentum loss and mass transfer, along with an 18.376-yr modulation interpreted as a light-travel-time effect from a third body. The TESS timings reveal anti-correlated variations between the primary and secondary minima that appear to result from differential rotation of high-latitude starspots. These results suggest that BS Cas is a physically complex contact binary in which long-term photometric and eclipse-timing variations reflect multiple processes, including magnetic activity and the influence of a third body.

We investigate topological superconductivity in a heterostructure consisting of a two-dimensional $s$-wave superconductor and a $d$-wave altermagnet with Rashba spin-orbit coupling. In particular, we study the effects of strain and hopping anisotropy on the topological superconducting phases. By calculating the Chern number, we obtain topological phase diagrams as functions of the chemical potential and the strength of the effective magnetic field induced by the magnetic proximity effect. We find that strain-induced lattice distortion allows topological superconducting phases to emerge over a broad parameter region. Our findings suggest that strain that lowers symmetry can serve as a route to realizing topological superconductivity in altermagnet-superconductor heterostructures.

Many temporal process learning and monitoring pipelines operate in local windows, making window-level decisions unavoidable in practice. In such settings, classical statistical tests can be applied to individual windows, but they typically evaluate predefined parametric hypotheses-such as unit-root or moment-based conditions-thereby limiting flexibility when reference behavior is defined empirically from task- or domain-specific data. In this work, we view window-level monitoring as a process control problem and reformulate it as reference-based hypothesis testing, where the null hypothesis is specified by an empirical reference distribution rather than a fixed parametric model. We operationalize this perspective through a representation-based, nonparametric framework that combines pretrained time series encoders, kernel density estimation, and conformal calibration, yielding finite-sample valid inference in learned representation space. Classical notions such as stationarity and cyclostationarity arise as natural instantiations of empirical reference sets within this framework. Through experiments, we demonstrate sensitivity to window-level distributional deviations while maintaining well-calibrated inference under stable reference regimes, highlighting the applicability of the proposed approach to a broad class of time series process control and monitoring tasks.

Choosing features is difficult with few labeled molecules, even for a tabular foundation model that learns in context. We study this problem with TabPFN for low-data molecular property prediction, separating input width, feature meaning, and predictive usefulness while keeping contexts, query molecules, and predictor settings matched. On ADME/Tox assays from Therapeutics Data Commons, endpoint-matched assay predictions improve a compact RDKit descriptor baseline, and a provenance audit shows that most audited query molecules already occur in the predictors' training sources. Across 77 FS-Mol development tasks and 157 held-out test assays drawn from ChEMBL, broad RDKit descriptor and fingerprint families provide a strong default, rank-averaging TabPFN with support-fitted fingerprint forests improves upon both models, and learned weighting of family predictions adds further gains at a larger prediction budget. A utility-trained selector builds a traceable 16-column input from 238 sourceassay predictors, improving upon the baseline while trailing wider representations and equalwidth structural SVD coordinates. Named outputs identify what each feature predicts and where it comes from. Their usefulness for a new assay is assessed alongside the other input features, and paired removal and replacement tests measure what each selected output contributes. Together, these studies offer a protocol for evaluating feature selection, representation choice, and data provenance with a fixed tabular foundation model for few-shot molecular property prediction.

Moire superlattices in transition metal dichalcogenides provide a highly tunable platform for exploring strongly correlated electronic phases, such as generalized Wigner crystals. While these crystalline states typically melt with increasing thermal fluctuations, an electronic analogue of the Pomeranchuk effect can stabilize the localized solid phase at elevated temperature through isospin entropy. Here, we report the observation of an electronic Pomeranchuk effect at the fractional filling factors of $\nu = 1/3$ and $\nu = 1/4$ in AB-stacked twisted bilayer MoS$_2$ with twist angles of 4.1$^\circ$ and 3.9$^\circ$, respectively. At ultra-low temperature, the system exhibits a highly conducting, itinerant behavior at these fractional fillings, characteristic of a compressible Fermi-liquid ground state. Upon heating, the system exhibits a counterintuitive increase in the longitudinal resistivity, signaling an isospin-entropy-driven transition into a localized generalized Wigner crystal state. Our findings highlight the unique capacity of flat bands in twisted bilayer MoS$_2$ for stabilizing highly degenerate magnetic configurations, offering new insights into the thermodynamic phase diagrams of low-dimensional correlated systems.

Data-driven design of frequency-selective surfaces (FSSs) requires reusable datasets that link structural geometry to electromagnetic response under documented simulation conditions. Here we present FSSDataBase, an openly available collection of 5,000 procedurally generated single- and multilayer FSS unit cells simulated using Ansys HFSS. The dataset covers 10-20 GHz with transverse-electric and transverse-magnetic relfection and transmission responses, including magnitude and phase, for recorded incidence angles of $0^{\circ}$ and $30^{\circ}$. Each record links binary structural masks and JSON-based reconstruction metadata to raw response samples and derived magnitude labels. Structural descriptors, response-distance measures and task-oriented cost functions characterize geometric and electromagnetic variation and support screening for polarization stability, angular-selective bandwidth and matched-amplitude, 180{\deg} phase-separated structure pairs. Accompanying code supports model reconstruction, data processing and configuration-driven generation of additional samples. Beyond conditional forward modelling and response-based structure retrieval, FSSDataBase has the potential to serve as a reference dataset for deep-learning-based FSS inverse design, supporting fair comparisons across network architectures under shared data splits,input conditions and evaluation protocols.

We consider the Korteweg-de Vries (KdV) approximation for the good Boussinesq equation. Building upon the low-regularity local-in-time justification established by Hong and Yang \cite{HY2024}, we prove that this approximation remains valid on logarithmically long time intervals. The key ingredient is a persistence of regularity argument, which allows the local approximation to be iterated using the rescaled conservation law for the Boussinesq equation.

GPU-dense AI data centers need to run on liquid cooling as air simply cannot shed the heat at these power densities. Yet the cooling loops themselves are blind to what workloads are about to run; they crank up flow only after a sensor catches a temperature climb, which can take 30 to 50 seconds. We built Job-Class Thermal Intent (JCTI) to close that window. The scheduler already knows a job is coming and what class it belongs to; JCTI feeds that information straight to the cooling controller so it can stage coolant before the heat shows up. We pulled the thermal signatures for each job class out of MLPerf GPU power traces and tuned arrival patterns against Alibaba cluster data. Over 120 paired Monte Carlo trials the numbers come out to 56.4% fewer thermal violations and 60.2% less cumulative overshoot than a straight PI loop. As AI data centers evolving towards gigawatt grid loads with highly fluctuating power swings, thermally-aware scheduling reduces sudden demand and improves load prediction in grid side. Cooling and scheduling have been running as two separate systems for years despite each one knowing something the other needs, JCTI wires them together.

This paper proposes an architecture for equipping large language models (LLMs) with audio-understanding capabilities without fine-tuning their weights. The proposed symbiotic architecture employs an injector module that writes audio-conditioned vectors directly into the target LLM's short-term memory, i.e., the key-value (KV) cache, enabling the LLM to behave as an audio language model (ALM). The architectural advantages are twofold. First, it improves the scalability of ALMs: because the proposed method bypasses the LLM during audio injection, the injection cost is governed by the injector width rather than the backbone width, and can therefore scale more slowly than the cost of full-backbone prefilling. Second, since the training scheme does not update the LLM weights, the original capabilities of the LLM are preserved without the risk of degradation from fine-tuning. The effectiveness of the proposed method is evaluated on both audio-understanding tasks (automatic speech recognition, audio question answering, and acoustic scene classification) and text-only tasks. We confirm that, while activating fewer parameters during audio prefilling, our architecture outperforms the conventional method with a frozen LLM and approaches the performance of a fine-tuned ALM, all while preserving the backbone LLM's original text-only task performance by construction.

Reasoning segmentation aims to interpret implicit textual queries and enable fine-grained visual perception, which is critical for applications such as human-computer interaction and embodied agents. Existing methods typically generate explicit Chain-of-Thought (CoT) by multimodal large language models (MLLMs) before localizing the target. Although intuitive, such explicit verbal reasoning introduces substantial attention interference: redundant textual tokens disrupt attention during perception-token generation and also increase the effective distance between visual tokens. To address this issue, we propose LIRSeg, which fully replaces explicit CoT with a compact set of learnable latent tokens for reasoning segmentation. LIRSeg is trained in two stages: spatial alignment grounds the latent tokens in object-relevant visual evidence, and GRPO further optimizes them with segmentation rewards. To make these compact latent tokens more informative, we introduce three complementary mechanisms from an information perspective: extreme-advantage sampling for selecting informative training signals, decoupled exploration-stability updates for learning complementary representations, and latent diversity amplification for preventing representational collapse. Extensive experiments on benchmarks demonstrate that LIRSeg consistently improves both segmentation accuracy and reasoning efficiency. Compared with the VisionReasoner baseline, LIRSeg achieves absolute gIoU improvements of 4.9% on ReasonSeg, 7.1% on MUSE, and 4.7% on MMR, while achieving a approximately 16x reduction in reasoning tokens. Code is available in supplementary materials.

We present three-dimensional smoothed particle hydrodynamics simulations of highly misaligned circumbinary disks (CBDs) around moderately eccentric equal-mass binaries ($e_\mathrm{b}=0.5$). We show that the binary accretion is modulated on the binary orbital period and exhibits two pulses near periastron. The dominant pulse peaks before periastron for an initial binary-disk misalignment of $60^\circ$, shifts to after periastron at $90^\circ$, and occurs at an even later post-periastron phase at $120^\circ$. We further show that the two pulses are accompanied by a time-dependent response of the circumstellar disks (CSDs) and by different distributions of accreting material within the cavity and around the CSDs. The qualitative pre- versus post-periastron distinction is also present in individual binary orbits despite variations in pulse amplitude. Our results motivate future tests of whether pulse timing is related to binary-disk orientation.

Clinical measurements are recorded for some patients but not others, at rates that differ across hospitals, and marginal conformal coverage does not ensure coverage within groups defined by missingness. We propose a missingness-aware conformal calibration procedure for mortality prediction under cross-hospital distribution shift. It selects a measurement on an independent sample, groups patients by whether that measurement is recorded, and applies Mondrian calibration within each group, so no calibration outcome is reused. We evaluate the procedure across hospitals in eICU and across care units within one MIMIC-IV hospital, using three predictors. Relative to pooled calibration, it reduces the average worst-group coverage gap on its selected groups in all six settings, with a median reduction of 1.9 percentage points; paired site-bootstrap intervals exclude zero in five. These gains do not extend uniformly. Calibration by predicted risk achieves smaller gaps on a broader panel of missingness groups, and when eICU hospitals are evaluated separately, the gain shrinks for all three predictors and reverses in sign for one. We explain this discrepancy with a hospital-level decomposition. Pooling reweights hospitals through a covariance between group shares and coverage errors, and lets errors of opposite sign cancel: weighting explains the reversal, and cancellation accounts for most of the attenuation for the other two predictors. Constructed population distributions show that pooled and within-hospital evaluations can rank calibration methods oppositely even without sampling noise. Pooled improvement alone therefore cannot establish better coverage within hospitals, even when the calibration groups are fixed.

Anthropomorphic artificial intelligence systems increasingly remember personal details, display empathy, and are engaged with as social counterparts, creating forms of risk that emerge from the evolution of the user-system relationship over time. Existing safeguards largely operate at the level of individual conversational turns and cannot determine whether a sequence of seemingly acceptable interactions is cumulatively moving a particular user toward harm. This paper introduces personalized trajectory-level safety, a framework that treats relational safety as a sequential decision problem over a latent escalation state inferred from the user's messages and influenced by the system's responses. At each turn, a screening step first discards any response strategy that does not preserve at least one safe continuation of the interaction under every plausible model of the user. Among the remaining strategies, we formulate action selection as multimodal trajectory sampling, and use a Generative Flow Network to generate diverse future evolutions in proportion to their plausibility, safety, and utility. The system then selects the strategy that preserves the largest fraction of safe and useful continuations. We evaluate the framework in simulation, calibrated on statistics reported for real human-chatbot interactions, and using response strategies derived from public benchmarks. Results show that trajectory-aware decision making substantially reduces the frequency of harmful states while keeping helpful interaction. This work reframes safety for anthropomorphic AI from response-level filtering to personalized control over the future evolution of human-AI relationships. The source code is available at https://github.com/benedettapicano/ANTHROPOMORPHIC_SAFETY_TRAJ.

Let $\mathcal{R}$ be the hyperfinite $\mathrm{II}_1$ factor. Considering the partial sum operators $(\mathcal{S}_n)_{n\geq 1}$ of the noncommutative Vilenkin-Fourier series associated with an arbitrary admissible Vilenkin group, we prove that there exists a universal constant $c>0$ such that \begin{equation*} \sup_{n\geq1}\|\mathcal{S}_n(f)\|_{L_{1,\infty}(\mathcal{R})} \leq c\|f\|_{L_1(\mathcal{R})},\quad f \in L_1(\mathcal{R}), \end{equation*} and, for every $1<p<\infty$, $$\sup_{n\geq1}\|\mathcal{S}_n(f)\|_{L_p(\mathcal{R})} \leq c\frac{p}{p-1}\|f\|_{L_p(\mathcal{R})},\quad f \in L_p(\mathcal{R}).$$ Besides the transference technique, the main novel ingredient is a modified version of noncommutative Calderón-Zygmund decomposition established in \cite{CCP2022}. Consequently, we resolve the problem of weak type $(1,1)$ estimate communicated to the authors by Fedor Sukochev, and substantially improve the strong type (p,p) estimates obtained in \cite{DFdePS2001} by achieving the optimal order $\frac{p}{p-1}$.

Let $\lambda$ denote the ratio of the length of a longest common subsequence of two length-$n$ strings to $n$. Rubinstein, Seddighin, Song and Sun [RSSS19] gave an $\Omega(\lambda^3)$-approximation for LCS running in $\widetilde O(n^{39/20})$ time, where $39/20=1.95$. Song [Son19] mentioned that improving the $n^{1.95}$ running time is an interesting open question. We give an algorithm that computes an $\Omega(\lambda^3)$-approximation of the longest common subsequence in $n^{8/5+o(1)}$ time. This improves the exponent $1.95$ to $1.6+o(1)$.

Moving Horizon Estimation (MHE) is a state estimation method based on finite-horizon optimization that can offer higher accuracy at the cost of increased computation compared to Kalman filter-based approaches. We present a linear smoothing MHE formulation as a dense Quadratic Program (QP), and a solver consisting of a continuous-time Newton's method augmented with the $\mathcal{L}_1$ Adaptive Optimizer ($\mathcal{L}_1$-AO). While MHE is inherently time-varying, conventional approaches treat it as a sequence of independent, time-invariant problems and employ iterative solvers at each time step, which can be both inaccurate and computationally burdensome. In contrast, time-varying solvers track the optimal solution with fewer iterations by exploiting the temporal evolution of the problem, thereby reducing the computational load. In this research, we enhance both the performance and efficiency of MHE through a time-varying solver with an $\mathcal{L}_1$-AO augmentation that compensates for the prediction inaccuracy, which is common in practice due to noisy sensors and the lack of prior knowledge of the system. Simulation results on a quadrotor platform show that the $\mathcal{L}_1$-AO-augmented approach solves the MHE optimization problem more efficiently than the baseline time-invariant solver and achieves higher estimation accuracy under challenging conditions, compared with both the Extended Kalman Filter and the standard MHE.

The fundamental competition between crack-tip cleavage and dislocation emission dictates the ductile-to-brittle transition in crystalline solids. In hydrogen-charged environments, this delicate balance is disrupted, yet existing models often struggle to self-consistently capture the atomic-scale thermomechanical modulation of the emission barrier. Here, we propose the hydrogen-informed Rice-Beltz framework to quantify the critical stress intensity factors (SIFs) required for initial dislocation nucleation under mixed-mode (I+II) loading. By integrating the elastic interaction of the dilatational field of interstitial hydrogen and the applied stress fields against the incipient dislocation core, the model recovers the non-monotonic relation between the most probable SIF and the hydrogen concentration. Unlike the classical pure mode-I (or mode-II) scenario, the minimization of the local strain energy density is employed to generate closed-form nominal driving forces, which are subsequently embedded into a transition-state-theory framework. This atomistically-informed approach yields the most probable SIF for the first dislocation emission event as a function of the loading rate, slip angle, and hydrogen concentration, providing a rigorous, parameter-free boundary condition for the onset of hydrogen-modulated crack-tip plasticity.

Detector simulation and event reconstruction will be computationally expensive for future $e^+e^-$ collider programs and are currently critical bottlenecks for accurate feasibility studies. To address this challenge, we build a Parnassus model for the CLD detector concept. Parnassus is a framework for automatically tuning a surrogate model, in our case, a conditional flow matching neural network, to emulate a full detector simulation and reconstruction. We train on $e^+e^-\to Z\to q\bar q$ events at $\sqrt{s}=91.2$ GeV processed through a Geant4 simulation of the CLD detector concept and the Pandora particle-flow reconstruction and reproduce single-particle kinematics, particle-IDs, and impact-parameter distributions. We also examine jet- and event-level features, inclusively and split by flavor, and find excellent fidelity, significantly better than the parameterized program Delphes. Furthermore, we train a transformer-based flavor tagger on the reconstructed particle-flow constituents and show that the surrogate preserves the $b/c/s/q$ discrimination of the full CLD reconstruction. The Parnassus CLD model achieves a generation cost of about 1.2 ms (40 ms) per event on a single GPU (CPU), over three (two) orders of magnitude faster than full simulation and reconstruction. Our model is publicly available for feasibility and design studies.

In face-to-face, real-time communication, a talking agent must track the target speaker through natural pauses, turn-taking, and backchannels, often amid background cross-talk. Most target speaker extraction (TSE) studies, however, rely on simulated mixtures with full or sparse overlap and ignore the turn-taking of real conversations. We therefore introduce, to our knowledge, the first benchmark for online audio-visual TSE (AV-TSE), built from intact dyadic interactions with independent third-party interference. Observing that anticipating upcoming activity from semantic, acoustic, and facial cues benefits online AV-TSE, we propose an LLM-based target-speaker voice activity projection (TS-VAP) module. Unlike conventional VAP with separated speaker channels, it forecasts the future activity of the target and conversational partner directly from the overlapping mixture, drawing on the linguistic and conversational knowledge of a speech-LLM, and uses this prediction to guide a low-latency separator. We further combine this predictive context with historical and synchronous speaker context. Experiments show that TS-VAP consistently improves streaming extraction across multiple AV-TSE backbones, and that further combining historical, synchronous, and predictive context yields nearly 1 dB gain on real AV conversations. Project page: https://jjjjiaozi.github.io/TS-VAP/.

Tandem speech-to-speech architectures couple a responsive speech frontend with an asynchronous text backend. In KAME, a large language model (LLM) serves as the backend, supplying candidate responses as guidance to the speech frontend while the user is still speaking. Ordinary conversation recordings capture the eventual response but not the guidance the backend would supply during the user's utterance. Generating the missing guidance with a simulator LLM adds substantial data-preparation overhead when training on real conversations. We propose randomized intermediate guidance, which derives guidance directly from the conversation corpus rather than simulating backend LLM behavior. During training, target responses provide informative guidance, while randomly sampled responses provide potentially irrelevant updates during the utterance. This combination aims to teach the frontend to use backend information selectively. On synthetic dialogues, KAME trained with this recipe achieves response quality comparable to that of the LLM-generated and similarity-based baselines. Training on 3.8k hours of real conversations improves smooth turn-taking and audio-judge naturalness over synthetic-data KAME while retaining a response-quality advantage over Moshi. These results show that randomized guidance offers a practical route to combining the response-quality benefits of tandem models with natural conversational behavior learned from real speech.

Spin-orbit coupling and electron correlation play important roles in a broad range of chemical and physical phenomena. In this work, we systematically assess the performance of single-reference algebraic diagrammatic construction combined with state interaction (SI-ADC) for describing electronic structure in the presence of spin-orbit coupling. We find that second-order SI-ADC methods (SI-ADC(2)) achieve zero-field-splitting accuracies comparable to those of higher-level and more computationally demanding approaches, including variational four-component ADC and two-component ADC methods based on spin-orbit effective Hamiltonians. We further demonstrate the broad applicability of SI-ADC by evaluating a diverse range of spin-orbit-coupled properties, including electron affinities, ionization potentials, neutral excitation energies, core-ionization energies, and magnetic g-tensors, for systems ranging from small molecules to transition-metal complexes and f-block compounds. Across these applications, SI-ADC(2) methods provide consistently accurate results, likely benefiting in part from favorable error cancellation between electron-correlation and spin-orbit effects. In contrast, SI-ADC(3) methods are generally less accurate, particularly for systems containing transition-metal and heavy elements. Overall, these results establish SI-ADC as an efficient and broadly applicable framework for incorporating spin-orbit coupling into ADC calculations of electronic and spectroscopic properties.

For $a$ in $\mathbb{R}$, a subset $V$ contained in $\mathbb{R}^{n}$ is said to be $a$-convex if $x, y \in V \Longrightarrow a x + (1-a) y \in V$. According to Pinch [Math. Proc. Cambridge Philos. Soc., 1985], the $a$-convex hull of $V$ is the intersection of all of the $a$-convex subsets of $\mathbb{R}^{n}$ that contain $V$, and Pinch also defines $D(a)$ as the $a$-convex hull of $\{ 0, 1 \}$ in $\mathbb{R}^{1}$. Pinch conjectured that if $a$ is a totally real algebraic integer and $D(a)$ has no limit points, then every algebraic conjugate of $a$ other than $a$ is in $(0, 1)$. We succeed in proving this conjecture, which seems to have remained open.