Papers

In magicians' forcing, spectators may experience a choice as self-determined even when that choice has been externally directed. Research on the sense of agency has developed largely around action-outcome relations, most notably intentional binding; however, the problem of choice authorship (why a particular choice is experienced as originating from oneself) must be treated as distinct. This paper integrates research on agency and forcing by distinguishing the locus of intervention within the choice-action-outcome chain, self-attribution at the Decision, Action, and Outcome levels, and the processes by which feeling of agency and judgment of agency are constructed. On this basis, a distinction is drawn between intentional binding, which concerns the temporal/causal relation between action and outcome, and the different binding problem exposed by forcing: the relation between a choice and its author. This theoretical relation is termed authorship binding. Authorship binding does not denote a new implicit measure; rather, it refers to the constructive relation through which a choice whose formation has been directed by external factors can nevertheless be experienced as originating from the self. Forcing can sustain this relation not by eliminating agency, but by selectively preserving, substituting, and redistributing agency cues. XR, in which body-centred spatial relations can be manipulated, is further conceptualised as a medium for extending the forced-yet-free structure into space.

The Kepler space telescope was a successful mission that used the transit method. Multi-planetary systems are interesting because they are dynamically rich due to the planets' interactions. Kepler's mission was the first to find a system with four planets all in resonance, that is Kepler-223. The Kepler-223 system is of unique dynamical interest as the first exoplanetary system confirmed to host four sub-Neptune planets in a precise 3:4:6:8 resonant chain. Previous dynamical studies have struggled to fully recover the long-term phase protection mechanisms, often finding that higher-order resonant angles circulate rather than librate. In this work, we re-examine the dynamical state of Kepler-223 by performing high-resolution N-body integrations and constructing detailed stability maps in the (a,e) phase space, we demonstrate that the system is locked in a deeper resonant state than previously thought. Crucially, we find that all resonant arguments, exhibit robust libration over secular timescales. Our stability maps reveal that the planets reside in clearly defined stability islands that allow for higher eccentricities than previous estimates. We further compare these numerical islands with analytical resonance widths and find excellent agreement for the inner three planets, while the outermost planet (Kepler-223e) exhibits a stability region wider than first-order analytical theories predict, suggesting complex multi-body stabilization effects at the edge of the chain.

We investigate a class of differential inclusions governed by non-autonomous and autonomous maximal monotone operators, involving set-valued perturbations with a Volterra integral term and subject to a nonlocal condition. Under suitable assumptions on the governing operator, the set-valued perturbation, and the Volterra kernel, we establish existence results for solutions. In particular, the set-valued perturbation is assumed to satisfy a one-sided Lipschitz condition, while the Volterra kernel is required to be Lipschitz continuous. The existence of a trajectory is established by means of an iterative construction and an application of Zorn's lemma. Finally, several examples are presented to illustrate the applicability of the abstract results.

Alignment-free RGB-T salient object detection (RGB-T SOD) aims to identify salient objects from unregistered RGB and thermal image pairs without costly pre-alignment. However, spatial misalignment breaks pixel-wise correspondence and causes feature contamination during cross-modal fusion. To address this issue, we propose S2A, a semantic-to-spatial alignment framework for alignment-free RGB-T SOD. Specifically, a global-guided hierarchical fusion module (GGHF) first exploits global semantic guidance to suppress background interference and refine hierarchical intra-modal features. Subsequently, the alignment-free cross-modal channel attention module (AFCA) globally exchanges complementary semantic information through channel-wise interaction, effectively overcoming the interference caused by local spatial misalignments. Finally, a spatial deformable cross-attention module (SDCA) predicts adaptive sampling offsets to recover local cross-modal spatial correspondence. Through this semantic-to-spatial paradigm, S2A first enables reliable cross-modal semantic interaction and subsequently performs local spatial calibration, effectively reducing misalignment-induced feature contamination. Without bells and whistles, S2A achieves highly competitive performance on multiple public alignment-free RGB-T benchmarks, demonstrating its effectiveness in alleviating misalignment-induced feature contamination.

In this paper, we develop a bordism-theoretic approach to rigidity problems for toric and quasitoric manifolds. We prove that two toric manifolds are isomorphic as varieties if and only if they are weakly equivariantly unitary bordant. We also establish a parallel rigidity result for omnioriented quasitoric manifolds satisfying the injectivity condition, showing that their equivariant unitary bordism classes completely determine their omniorientation-preserving equivariant homeomorphism types. Thus, equivariant bordism provides a topological framework for detecting geometric and combinatorial rigidity.

Machine fault diagnosis from vibration requires learning from scarce labelled fault recordings while meeting the computational constraints of edge devices for local inference. We introduce DualRes, a compact oscillatory state-space model that combines two complementary spectral views of vibration, capturing rapid changes and fine frequency structure. Time-aligned views are processed by selective oscillatory memory, which learns how long to retain temporal patterns. The encoder contains 39,528 parameters. We evaluate supervised learning across six bearing datasets and a gearbox benchmark, with an additional gearbox pilot. Recording-level splits and explicit accounting of labelled duration distinguish data efficiency from repeated exposure to correlated samples. On the main gearbox benchmark, DualRes achieves state-of-the-art performance among the nine evaluated methods at six of seven label budgets. With about six labelled seconds per class, it improves macro-F1 by 16.1 percentage points over the next strongest comparator. On the same benchmark, DualRes achieves a 1.44-fold recording-level speedup and a 24.8-fold reduction in checkpoint storage relative to a selective state-space baseline under matched hardware and runtime conditions. Bearing results reveal task-dependent trade-offs. These findings support oscillatory memory as a compact approach to vibration diagnosis under limited labelled exposure.

The estimation of the breathing mode frequency for viral capsids remains a persistent challenge across literature which spawned various frameworks and methodologies. However, discrepancies were observed between the experimental Low Frequency Raman Scattering (LFRS) and calculated pre-existing values. To solve this discrepancy, we propose a framework which is developed to determine the breathing-mode frequency by modelling the virus as a macroscopic coupled harmonic oscillator. By integrating mass-loading directly into the classical elastodynamic equations, this model yields an analytical expression that couples the system's total inertia with its specific geometry. Moreover, the viruses are segregated according to their geometries into three coordinates to acquire their respective geometric eigen values. Here we illustrate how the proposed Multigeometric Breathing mode Framework for Virus (MBFV) outperforms prior models by yielding closer values than the pre-existing frameworks upon validating against LFRS.

Limited expert annotation capacity is a pervasive constraint in biodiversity monitoring. Passive acoustic recorders and camera traps generate data faster than experts can analyse them. Machine learning (ML) models can process these data at scale, but their reliability depends on the quality, quantity, and coverage of labelled samples, so expert time remains a constraint. Active learning (AL) eases this bottleneck by selecting, under a fixed annotation budget, the samples expected to improve a model most, and published evidence shows it can reduce the labels needed to reach a target performance. Monitoring programmes, however, face a broader question: how should a limited expert budget be divided so that model training, validation, and the ecological estimates built on model outputs all remain reliable? Because AL selects samples non-randomly, its labels are unsuitable for validation, calibration, or threshold selection, a tension rarely acknowledged. We synthesise AL research across acoustic and image modalities and identify gaps and opportunities. Most studies evaluate query strategies on pre-labelled benchmarks with simulated annotators; deployments in real monitoring workflows are rare and concentrate on birds and cetaceans. Bats, insects, amphibians, and fish are underrepresented, and multimodal applications remain largely unexplored. Evaluation centres on headline reductions in annotation effort, often without random-sampling baselines, per-class results, or calibration analysis, and rarely accounts for the labels required for validation. We provide a tutorial treatment of the AL loop that makes these budget decisions explicit, and a roadmap towards AL methods that support label-efficient training, validation, and trustworthy downstream ecological inference.

Benchmarks for vision-language models offer their answer choices in some convention: a letter, a color name, a pixel coordinate. That convention is treated as neutral. We find it is not, and that the limits a benchmark reports can belong to the readout rather than to the model. On 200 COCO photographs, Qwen3-VL-4B picks the correct one of nine locations for a named object 68.5% of the time when the locations are given in English and 20.0% when the same locations are given as pixel coordinates. Chance is 11.1%. The cost arises when the answer options are coordinates; giving the model a coordinate in the question instead costs 3.5 points and is not significant. The gap holds on a 4x4 grid, under 8-bit rather than 4-bit quantization, and in every slice by object size, boundary distance and category. It also decides which model wins. Two models that tie under English names differ by 39 points in one coordinate system and by 54 in the other, in opposite directions. On the color task, three of the four open models capable of the task show the penalty; on photographs, two of three open models do, and so does Gemini, at 11.1 points on parseable answers (p = 1e-4). GPT-4o does not. To ask whether a model reads a coordinate at all, we attach the wrong name to each one and record which the model follows. Color options written as hue angles are followed below chance; a normalized pixel convention is followed at four times chance. This tells apart conventions a model can use from ones it cannot, though it did not predict accuracy on two untried conventions. Five models also name the same color wheel five different ways, so a fixed answer vocabulary is not neutral across models either. Five times during this work we measured a capable model as incapable because our scorer and the model disagreed about what an answer looks like. We report each case. They are the phenomenon in miniature.

Blockade, a phenomenon in which one particle hinders the transport of another, constitutes a core research topic in mesoscopic and microscopic physics. Benefiting from the unique single-particle characteristics, it underpins versatile applications spanning from quantum computation to quantum metrology. Nevertheless, in the context of quantum energy?storage devices, i.e., quantum batteries (QBs), blockade triggered by the intrinsic level splitting is in turn detrimental to energy accumulation. Here we propose an antiblockade charging protocol enabled by deliberately engineering a bright-dark interaction. Rooted in a manifold of equally-spaced dressed-dark states, this elaborate configuration opens up a new charging channel facilitating an antiblockade energy flow and outperforms the Rabi-split blockade design with at least a 20-fold enhancement in energy storage. Additionally, we further illustrate the achievement of superextensive (quadratic) charging characteristic by operating the system in the bad-cavity limit. In view of its inherent generality, our scheme offers a universal and feasible pathway toward high-performance quantum energy architectures.

Deploying large language models (LLMs) on sensitive data via Retrieval-Augmented Generation (RAG) introduces severe privacy risks. Recent studies apply Differential Privacy (DP) to LLMs with RAG for formal privacy guarantees. However, existing DP-RAG frameworks rapidly exhaust the privacy budget. Although recent efforts attempt to save the budget by narrowing the retrieval scope or sparsifying private generation, these methods themselves cumulatively consume the budget, whereas they could actually rely merely on public information or at a negligible one-time privacy cost. This mismatch fails to align budget expenditure with the model's actual reliance on private data, causing substantial waste on operations that require no private access. To address this, we propose SparsePay-RAG, adopting "only pay what you must spend" as its core principle. Using public information as a zero-privacy prior, it charges the privacy budget only for the private increment. Specifically, SparsePay-RAG narrows the retrieval scope via public topic-guided clustering, adaptively controls private access frequency without privacy cost through isotonic cross-layer trajectory fitting, and compresses per-access budget via DP contrastive decoding. Under strong privacy constraints, experiments show SparsePay-RAG achieves superior privacy-utility trade-offs over baselines.

The discovery of next-generation Co-based superalloys with improved high-temperature stability is hindered by the vast compositional design space and complex interactions among alloying elements governing gamma-prime phase stability. This study presents a physics-informed machine learning framework for the inverse design of Co-based superalloys with higher gamma-prime solvus temperature while accounting for alloy density. Four descriptors representing atomic size mismatch (delta-MV), mixing enthalpy (Delta-Hm), electronegativity mismatch (delta-EN), and valence electron concentration mismatch (delta-VEC) were used to characterize the chemistry governing phase stability. Regression algorithms were evaluated using leave-one-out cross-validation, with Gaussian Process Regression (GPR) achieving the best performance (R2 = 0.932, RMSE = 34.7 deg C, and MAE = 25.9 deg C). The probabilistic nature of GPR enabled uncertainty-aware Bayesian optimization for exploring the uncharted composition space. Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Genetic Algorithms were used to simultaneously increase gamma-prime solvus temperature and reduce alloy density. The optimization generated numerous previously unexplored Co-based alloy compositions, with predicted gamma-prime solvus temperatures of 1248-1353 deg C, including candidates approaching or exceeding the maximum value in the experimental dataset. Descriptor analysis identified atomic size mismatch and mixing enthalpy as key factors governing gamma-prime stability, while principal component analysis confirmed that the proposed alloys occupy chemically relevant descriptor regions. This physics-informed inverse design framework provides an efficient route for identifying high-performance Co-based superalloys and can be adapted to other advanced structural alloys.

Trust can attract fraud when it delays verification. We develop a two-market signaling model in which generative AI lowers fabrication, verification, and targeting costs. When fabrication becomes profitable before verification, claim credibility first falls and later recovers. Across markets, higher prior quality can delay verification, creating an interval in which only the lower-quality market checks. If targeting becomes profitable in this interval, deceptive sellers enter the higher-quality but less vigilant market, and their entry can initially reverse its reliability advantage. The inflow also triggers verification and deters further entry. We call this self-limiting mechanism trust arbitrage. In the age of generative AI, trust can thus create an endogenous but temporary protection gap that redirects deception across markets.

The $^{9}\mathrm{Be}(p,\alpha)^{6}\mathrm{Li}$ reaction is a destruction process for $^{9}\text{Be}$ in stars and the Big Bang. The Trojan Horse Method (THM) is a well-established indirect technique that enables the determination of reaction cross sections within the Gamow energy region while avoiding the uncertainties associated with low-energy extrapolations of the $S(E)$ factor and electron-screening effects. In this work, the bare-nucleus THM data reported by Wen et al. are systematically analyzed using polynomial fitting, a double Breit–Wigner model, and the $R$-matrix formalism. The applicability and physical implications of these theoretical approaches in describing the low-energy $S(E)$ factor of the $^{9}\mathrm{Be}(p,\alpha)^{6}\mathrm{Li}$ reaction are investigated and compared. Since the THM data have been incorporated into the NACRE II reaction-rate compilation, leading to improved accuracy in reaction-rate evaluations, the refined theoretical analysis presented here provides further insight into the underlying reaction dynamics and establishes a more reliable foundation for stellar reaction-rate calculations. The results obtained in this work offer valuable reference for future high-precision experimental investigations and the development of theoretical models in nuclear astrophysics.

The design of credible auctions—mechanisms where a revenue-maximizing auctioneer has no incentive to deviate from the protocol—faces a fundamental cryptographic barrier when the auctioneer controls shill bidders. While a natural approach is to use Secure Multi-Party Computation (MPC) to remove the trusted auctioneer, the impossibility of fair coin flipping of Cleve (1986) implies that monolithic MPC protocols grant the auctioneer a "free option": they can learn the auction's outcome and unilaterally abort if the revenue is unsatisfactory. Cryptographic commitments with ex-ante penalties mitigate this abort asymmetry, but no finite penalty suffices for heavy-tailed distributions. We circumvent this barrier by introducing the MPC Decomposition Principle. Rather than encrypting the entire mechanism, we use MPC strictly as an information-restriction tool. We isolate the winner determination problem into a minimal MPC gadget that computes and reveals the winner's identity but no payment information. This qualitative restriction mathematically bounds the information leaked upon an abort. By combining this gadget with sequential revelation and finite economic penalties, we design the Sequential Revelation Auction (SRA). We prove that bounding the information leakage strictly bounds the value of the free option, showing that a penalty of $k \geq \sum_{i=1}^n Rev(F_i)$ is sufficient for credibility, and tight: for equal-revenue distributions, every smaller penalty admits a profitable deviation. Using constant-round MPC, the SRA resolves an open question of Akbarpour and Li (2020) and Ferreira and Weinberg (2020) by providing a constant-round, incentive-compatible, revenue-optimal credible auction for all product distributions with vanishing revenue tails

In our previous results, we have proposed a new free boundary problem representing the bread baking process in a hot oven. The problem consists of conservation laws for the internal energy and water mass, the Stefan condition, the boundary and initial conditions. We note that the boundary condition for the water content contains the temperature at the boundary. Accordingly, we do not expect a strong solution for $W^{1,2}$-initial data. To address this difficulty, we established strong solvability by assuming stricter conditions for the initial functions in [3]. In contrast, this paper aims to prove the existence of a weak solution, when these assumptions for initial functions are relaxed.

Proximal-Tracking is a novel algorithm proposed in [Falsone and Prandini, Automatica, 135 (2022), 109938] for nonsmooth distributed consensus optimization with local set constraints. The global convergence and numerical behavior of Proximal-Tracking have been well studied in [Falsone and Prandini, Automatica, 135 (2022), 109938]. As a complement, in this paper we provide a theoretical analysis regarding its linear convergence rate under additional strong convexity and calmness assumptions. In particular, this analysis does not require assumptions such as differentiability or smoothness, thus preserving the original advantages of the algorithm.

Recent zero-shot text-to-speech (ZS-TTS) systems can reproduce a speaker's voice with high fidelity from only a few seconds of reference speech, raising concerns over unauthorized voice cloning and impersonation. Speaker identity unlearning has recently emerged as an approach to selectively suppress this capability for speakers who opt out while preserving synthesis capability for other speakers. Although existing approaches reduce speaker similarity, preventing re-identification often faces severe degradation of speech quality. Motivated by this observation, we propose GUARD, a lightweight speaker identity unlearning framework that combines a learned speaker gate with speaker-agnostic activation steering on a frozen TTS backbone. The steering vectors are optimized using group-relative reward optimization to shift outputs from forget speakers toward population-level impostor similarity while preserving intelligibility and speech naturalness. On CosyVoice2, GUARD reduces forget-speaker similarity from 0.541 to 0.103 and re-identification accuracy in a 150-speaker gallery from 73.5% to 0.5%, while preserving retain-speaker reproduction. The results demonstrate that similarity reduction alone may not fully characterize successful speaker identity unlearning and highlight re-identification as a complementary criterion for its evaluation.

Molecular imaging has been transformed by integrating positron emission tomography (PET) or single-photon emission computed tomography (SPECT) with x-ray computed tomography (CT). However, x-ray CT in hybrid imaging is used primarily for anatomical localization and for attenuation and scatter correction of emission data. Its ability to characterize tissue composition remains underutilized. Spectral CT, including dual-energy CT (DECT) and photon-counting CT (PCCT), can provide material-specific information, such as images of iodine concentration, bone or calcium fractions, electron-density, effective atomic number and other spatially-varying material properties. These capabilities create an opportunity to bring tissue-composition information into molecular imaging. In addition, spectral CT can also reduce CT artifacts and provide more accurate corrections for emission data. In this review, we discuss the technical basis, integration pathways, and translational opportunities for combining spectral CT with molecular imaging. We review sequential and integrated PET/DECT approaches, emerging PET/PCCT concepts, PET-enabled spectral CT, and extensions to SPECT/CT. We then examine how spectral CT information may improve attenuation correction, scatter correction, tissue-fraction correction, positron range correction, reconstruction priors, and dosimetry. Clinical and translational applications include contrast-enhanced molecular imaging, vascular and perfusion imaging, quantitative bone marrow imaging, musculoskeletal imaging, and theranostics. Overall, spectral CT integration may move hybrid molecular imaging toward tissue-composition-informed molecular imaging, in which radiotracer signals are interpreted within their material context.

Approximately half of the Fermi-LAT-detected blazars lack spectroscopic redshift measurements, which limits population studies and investigations of the cosmological evolution of the gamma-ray blazar population. We aim to develop a multi-model ensemble framework to systematically evaluate the performance of machine-learning models for blazar redshift estimation and to provide predictive redshift probability distributions with quantified uncertainties for all blazars without measured redshifts. We construct an ensemble of 35 machine-learning regression models in the $1/(1+z)$ space. Predictions from individual models are combined to build non-parametric redshift probability density functions (PDFs) for each source, allowing an explicit characterization of predictive uncertainty. The source-level PDFs are subsequently stacked to derive population-level redshift distributions. Across the diverse models within our framework, we find comparable predictive performance, indicating a feature-limited performance plateau. We compile a catalog of predictive redshift distributions for 1,590 blazars without measured redshifts. The stacked PDFs recover the characteristic double-peaked redshift structure of blazars without using source class labels. For blazar candidates of uncertain type, the predicted redshift distribution exhibits a bimodal structure, suggesting a slightly larger contribution from FSRQ-like sources. Our multi-model statistical framework achieves competitive predictive performance while retaining faint sources. We provide predicted redshift distributions with quantified uncertainties for 1,590 blazars without measured redshifts, offering improved support for population and evolutionary studies. More broadly, this approach highlights the practical value of uncertainty-aware multi-model regression ensembling in astronomical data analysis.

DSpark-style parallel drafters have made speculative decoding highly effective, yet their draft phase remains serialized on the critical path of every round. Parallel speculative decoding (PSD) overlaps drafting with verification, yet existing methods must guess the accepted prefix and bonus token in advance: a wrong guess reverts the whole batch to serial drafting. We present DPara, a PSD framework that reuses effective parallel drafters yet guarantees backbone–verification overlap in every round, thereby eliminating this probabilistic fallback altogether. While the target verifies, DPara's diffusion backbone precomputes draft representations for every acceptance boundary with the bonus left unspecified; a lightweight autoregressive head then combines the revealed verification outcome with the matching precomputed representation to emit the next round's draft tokens almost instantly—fully parallelizing the dominant backbone forward with verification and leaving only the negligible head cost serial. Experiments on Qwen3-8B and Qwen3-14B across seven math, coding, and chat benchmarks show that DPara achieves average speedups of $3.21\times$ and $3.52\times$ over autoregressive decoding, surpassing the strongest serial and parallel speculative decoding baselines alike.

Compact vision-language models (VLMs) now power a growing share of multimodal applications. The benchmarks used to compare them, however, inherit a frontier-centric design: each model is reduced to a single accuracy number, narrowing the inter-model gap on saturated suites and pressing models into low-score bands on harder ones. We introduce PRISM-VLM, a multi-axis discriminative benchmark that scores every item along seven axes covering the recurring failure modes (task quality, behavioral robustness, and capability bottlenecks) and combines them into a single PScore, with items recycled from fifteen public benchmarks. Across compact VLMs from the past two years, PScore separates model pairs more reliably than prior single-axis benchmarks under an item-level paired bootstrap, and surfaces behavioral differences these benchmarks average away. Even models with statistically indistinguishable PScores diverge sharply along the per-axis profile, particularly on sycophancy, which is nearly orthogonal to single-prompt accuracy. We will release the full pipeline, prompts, and per-item annotations.

Automotive spinning FMCW radar provides dense, $360^\circ$ sensing and remains reliable under poor illumination and adverse weather, making it well-suited to autonomous navigation. Place recognition uses these observations to identify previously visited locations for re-localization and long-term navigation. However, heading changes appear as circular shifts in the polar radar representation, and conventional global aggregation can lose relationships among radar responses that are important for distinguishing similar places. We propose SGCA-Net, a spinning radar place recognition framework that combines rotation-robust feature extraction with Spatially Gated Correlation Aggregation (SGCA). SGCA learns spatial weights to reduce the influence of unstable and ambiguous radar regions, while aggregating pairwise correlations among local responses to preserve informative feature relationships. Experiments on the MulRan dataset show that SGCA-Net consistently outperforms SOTA methods across urban, campus, and open-road environments, while remaining robust to substantial heading variation. Evaluation on the HeRCULES dataset further demonstrates that SGCA-Net generalizes to unseen environments and radar sensors without fine-tuning.

Basing on new curvature conditions for the Bochner technique of Petersen-Wrink in \cite{PW21}, in this paper, we give some vanishing properties for weighted $L^Q$ $p$-harmonic $\ell$-forms on weighted manifolds, where $p>1, Q>p-1.$ These results generalize some previous theorems for $L^Q$ harmonic $\ell$-forms of Dung-Dung-Hung on the paper \cite{DDH25}.

Rely/Guarantee is a well-known verification technique for reasoning about concurrent programs. However, for some algorithms, devising suitable rely and guarantee conditions is challenging, due to the strong interference exhibited in these algorithms. The Ben-Ari concurrent garbage collector is an algorithm where the complex interactions between the components prevent the construction of compositional rely and guarantee conditions. This paper investigates an approach for verifying the Ben-Ari algorithm, which enables reasoning to be performed in a more compositional manner in cases where compositional reasoning would not otherwise be possible. This is accomplished by reasoning that a given property holds for all instances of a particular variable and then instantiating the variable to the local variable required. As well as providing a reasoning approach which is more compositional, the result is the identification of the core property required of a component, thus enabling a deeper understanding about the reasons why the algorithm works correctly. This helps to reveal the reasons why the rely/guarantee approach does not work directly for some problems.

We report single crystal growth and physical property characterization of CaMn_{2+x}Al_{10-x} crystals grown using a fractionation approach. Three consecutive batches were obtained from the same initial melt composition, systematically sampling neighboring, narrow regions of the ternary Ca-Mn-Al phase diagram. Single-crystal X-ray diffraction measurements reveal Mn/Al site mixing, with the excess Mn concentration, x, decreasing systematically with decreasing decanting temperature. A low-temperature upturn in magnetization and a maximum in electrical resistivity both become progressively weaker with decreasing excess Mn content, suggesting that the previously reported magnetic properties attributed to stoichiometric CaMn_{2}Al_{10} instead arise from an inhomogeneous distribution of excess Mn atoms. More broadly, the results presented in this work demonstrate how fractionation can help distinguish intrinsic magnetic behavior from disorder-induced effects associated with an inhomogeneous distribution of magnetic ions in materials with a finite width of formation, particularly when itinerant magnetism is under consideration.

Long-range interacting quantum systems exhibit interesting phenomena beyond the nearest-neighbours regime. Such interactions can reach further separated particles and change the behaviour of many-body correlations. Here, we analyse their structure in the antiferromagnetic long-range interaction transverse field Ising model with power-law interactions decaying as $1/r^{\alpha}$. We examine the spatial behaviour of basis-independent total multipartite correlations, called sector lengths $S_k$, of its ground state in the vicinity of critical points for different interaction exponents $\alpha$ and fixed system size. These correlation functions were shown to indicate entanglement and the existence of local realistic models, allowing us to make claims about the Bell nonlocality and entanglement in the system. We further investigate them in the context of non-stabiliserness, often referred to as quantum magic, which holds information on their classical simulability and usefulness for quantum computation. We evaluate Rényi stabiliser entropy near and outside the criticality and show that cumulative magic, up to the critical point, increases with $\alpha$ for all of the examined system sizes. Our results provide new insights into the interplay between the long-range interactions, observable non-classical phenomena and their classical simulability.

Audio language models understand what is said far better than how it sounds. Closing this gap takes more than data. Detailed acoustic annotation is costly, labels from stronger models inherit their errors and limits, and fixed data cannot adapt as the learner improves. We therefore propose EvoAudio, a recursive self-improvement system for audio understanding. To our knowledge, it is the first to evolve the model, waveforms, questions, and difficulty in one closed loop. EvoAudio uses the current model's performance to set the focus and difficulty of the next training data. A library of audio tools then constructs questions whose answers follow from how the audio was made, providing verifiable supervision without new human annotation. Reinforcement learning updates the model, and validation decides whether it enters the next evolution round. Across 13 rounds, EvoAudio improves five models with different audio encoders and language backbones on MMSU, MMAU-Pro, and MMAR. It achieves the highest average for every backbone, raising overall performance by up to 6.3 points. The improvement unfolds over successive rounds, with each stronger model starting the next round.

The quantum anomalous Hall effect carries dissipationless chiral edge currents without an external magnetic field, yet raising its operating temperature and controlling the number of edge channels remain difficult. Using first-principles calculations, we identify the two-dimensional metal-organic kagome ferromagnet Eu$_2$(C$_6$H$_4$)$_3$ as an intrinsic Chern insulator whose topological gap can be enlarged by mechanical strain. The monolayer has a spin-orbit-coupling-induced gap of 72.7 meV that widens to 124.7 meV under -8% biaxial strain, realizing a quantum anomalous Hall phase with Chern number $\mathcal{C}=-1$ and a single chiral edge state. The local moments come from the half-filled Eu $4f^{7}$ shell, while the gap-opening spin-orbit coupling is carried by Eu $5d$ states hybridized into the carbon kagome bands. The AB-stacked bilayer couples ferromagnetically and accumulates the per-layer Chern numbers, giving $\mathcal{C}=-2$ with two co-propagating chiral channels. An out-of-plane electric field then drives the bilayer through a sequence of topological transitions among $\mathcal{C}=-2$, $-3$, and $-1$, switching the number of edge channels. Together, strain, stacking, and gating give three distinct handles on the gap and the Chern number within a single stoichiometric material.

AraGenre is a shared task on hierarchical, definition-guided Arabic genre classification, motivated by the limited availability of annotated data in Arabic and other low-resource languages. Systems assign each Arabic text segment both a broad communicative genre and a fine-grained specific genre. The released training and development sets contain limited, primarily synthetic and controlled examples, whereas the hidden final benchmark contains noisier naturally occurring text spanning Modern Standard Arabic, Classical Arabic, and multiple dialects. Participants received natural-language definitions for 74 previously unseen specific genres, creating a zero-shot label generalisation setting in which systems had to infer class semantics rather than memorise fixed label-feature associations. The task attracted 46 registrations and 373 submissions, with 17 teams completing the final evaluation. Thakaa ranked first with a Hierarchical Macro F1 of 0.7352, followed by HoangPhong (HP) with 0.7169 and NAMAA with 0.7013. The results show strong broad-genre recognition but a substantial gap in fine-grained classification under linguistic and domain variation.