Learning-based semantic communication (SemCom) has recently emerged as a promising paradigm for improving the transmission efficiency of wireless networks. However, existing methods typically rely on extensive end-to-end training, which is both inflexible and computationally expensive in dynamic wireless environments. Moreover, they fail to exploit redundancy across multiple transmissions of semantically similar content, limiting overall efficiency. To overcome these limitations, we propose a channel-aware generative adversarial network (GAN) inversion-based joint source-channel coding (CAGI-JSCC) framework that enables training-free SemCom by leveraging a pre-trained SemanticStyleGAN model. By explicitly incorporating wireless channel characteristics into the GAN inversion process, CAGI-JSCC adapts to varying channel conditions without additional training. Furthermore, we introduce a cache-enabled dynamic codebook (CDC) that caches disentangled semantic components at both the transmitter and receiver, allowing the system to reuse previously transmitted content. This semantic-level caching can continuously reduce redundant transmissions as experience accumulates. Extensive experiments on image transmission demonstrate the effectiveness of the proposed framework. In particular, our system achieves comparable perceptual quality with an average bandwidth compression ratio (BCR) of 1/224, and as low as 1/1024 for a single image, significantly outperforming baselines with a BCR of 1/128.

Three-dimensional seismic full-waveform inversion (FWI) provides high-fidelity subsurface velocity models but is restricted by high computational cost, strong nonlinearity, cycle-skipping, and heavy dependence on initial models. Although data-driven deep learning mitigates these issues, it still produces over-smoothed results with limited physical interpretability and low efficiency. To address these challenges, we propose a dual-domain sparse deep learning framework for 3D seismic velocity inversion using the discrete cosine transform (DCT). DCT compresses seismic data and velocity models into a sparse domain to remove redundancy while preserving key structural features. A geometry-adaptive network named SEDCN (Squeeze-and-Excitation Deformable Convolutional Network) is adopted to better capture irregular salt-dome geometries and sharp velocity boundaries. We train and validate the network on 676 samples from the 3D SEG/EAGE salt model, with two schemes for comparison: the proposed DCT-SEDCN and the baseline SEDCN without DCT. Numerical results show that DCT-SEDCN reduces training time by more than 90% and achieves higher PSNR and SSIM than conventional spatiotemporal-domain methods. It effectively suppresses over-smoothing, recovers salt body boundaries and stratigraphic details clearly, and generates geologically more reliable velocity models. This study confirms that DCT-based sparse representation combined with geometry-adaptive deep learning significantly improves the efficiency, accuracy, and robustness of 3D seismic velocity inversion. The framework offers a scalable solution for large-scale 3D FWI and can be extended to elastic/viscoelastic multi-parameter inversion and field data applications.

The development of high-power single crystal fiber (SCF) lasers is critically hindered by the lack of a reliable cladding scheme to confine the optical mode and ensure beam quality. Here, we propose and demonstrate a two-step tapering-collapse method for the first time to fabricate a high-quality cladding on Yb:YAG SCFs based on elemental interdiffusion. This in-situ formed crystalline transition layer with a graded refractive index effectively suppresses lattice mismatch and abruptly mitigates core-cladding interfacial stress. Consequently, the numerical aperture of the SCF is significantly reduced from 0.280 to 0.199. In a master oscillator power amplifier configuration, the clad SCF delivers a remarkable 46.7% enhancement in slope efficiency compared to its bare counterpart, accompanied by a substantially improved near-field beam profile. This work establishes a facile and effective route to high-performance clad SCFs, unlocking their full potential for next-generation extreme-condition lasers.

The dielectric response and structural properties of finite-temperature electron liquids are central to accurately describing the physical behavior of electronic systems. This study presents a robust analytical model for the static structure factor of the uniform electron gas, combining physically motivated form for the static structure factor with constraints derived from high-accuracy path integral Monte Carlo simulations. The model accurately reproduces key features of the static structure factor across a broad range of temperatures and densities. Using this static structure factor, the density response function is directly evaluated, enabling a self-consistent definition of the static local field correction. As practical applications, the model is employed to investigate the low-velocity stopping power and the electron-ion friction coefficient. Results derived for the friction coefficient show good agreement with simulation data at moderate coupling and degeneracy. The proposed approach provides a computationally efficient and reliable method for characterizing the static response properties of correlated electron systems, facilitating improved simulations of energy deposition and ionic transport in warm dense matter and other strongly coupled quantum plasmas.

Gravitational wave astronomy is rapidly advancing with the development of new observatories, leading to an increasing volume and complexity of data. This trend places growing pressure on classical data analysis methods and motivates the exploration of quantum approaches. In this work, we introduce a quantum matched filtering framework for gravitational-wave detection based on the Long algorithm, marking its first application to the gravitational-wave data analysis. Numerical simulations show that the proposed approach preserves the quadratic speedup of quantum search while exhibiting significantly improved robustness, thereby overcoming key limitations of the Grover-algorithm based methods.

This work explores the differences between static and dynamically evolving physico-chemical models of pre-stellar cores. A 3D MHD model of a pre-stellar core embedded in a dynamic star-forming cloud is post-processed using sequentially dust radiative transfer, a gas-grain chemical model, and a non-LTE line-radiative transfer model. The chemical evolution is modeled along $\sim$20,000 tracer particle trajectories to capture the impact of a realistic dynamical evolution as the core is formed. The emission morphology of CH$_3$OH and $c$-C$_3$H$_2$ and the intensities of CH$_3$OH, $c$-C$_3$H$_2$, CS, SO, HCN, HCO$^+$ and N$_2$H$^+$ are compared with observations of L1544. Our results show a distinct difference in chemical morphology between the dynamical and static models. The dynamical model reproduces the observed spatial distribution of CH$_3$OH and $c$-C$_3$H$_2$ toward L1544, whereas the static model fails to reproduce this morphology. In contrast, when comparing modeled and observed intensities across a broad range of molecules, the static model shows good agreement with observations for L1544. The dynamical model systematically predicts lower abundances and modeled intensities for six of the seven species presented here. For sulphur-bearing species, the intensities are in better agreement with observations when the initial abundances are undepleted in heavier elements. This study reveals distinct differences between dynamical and static physico-chemical models. The static model predicts higher abundances and intensities for the majority of the molecules studied here, compared with the dynamical model. This discrepancy may stem from the specific choices of initial conditions, which could limit the dynamical models ability to fully capture the physical and chemical history. The intensities predicted by the static model are comparable to those observed toward L1544.

A booklet is a geometric structure formed by gluing multiple bulk spacetimes along a common interface and imposing gravitational consistency conditions at the junction. We have systematically investigated the properties of booklet structures, constructed the booklet geometry, and derived the multiway junction conditions applicable at the interface. In this work, we provide a complete solution to the multiway junction conditions for booklets composed of bulks governed by JT gravity. By constructing invariants of the dilaton solution space, we classify all dilaton configurations into three inequivalent types, each exhibiting attractive, repulsive, or neutral behavior. Through continuous isometric transformations, each type is fixed to a standard form characterized by a single physical parameter, effectively eliminating redundant degrees of freedom. This process selects a distinguished class of Poincaré coordinates for each type. Expanding the constraint equations order by order breaks coordinate invariance beyond the leading and subleading orders. By jointly solving the junction and continuity conditions up to subleading order, we find that junctions are only allowed when a sufficient number of bulks carry attractive dilatons, as captured quantitatively by a equilibrium condition. We further analyze all possible combinations of different dilaton types and determine the shape of the interface along with the explicit form of the dilaton defined on it.

Chemical activity is known to affect phase coexistence and coarsening in liquid mixtures, most commonly through reaction-induced changes of intermolecular interactions. Here, we analyze a scenario in which chemical reactions regulate particle transport while leaving thermodynamic interactions unchanged. We study an incompressible mixture of thermodynamically identical solutes with unequal diffusivities that interconvert through driven chemical reactions. Using linear stability analysis and finite-element simulations, we show that the system can phase-separate into solute-rich and solute-poor domains via two qualitatively different pathways. When interactions are too weak to induce phase separation, patterns arise through a generalized mass-redistribution instability and coarsen uninterruptedly. When interactions favor phase separation, coarsening can be arrested if chemical activity locally enriches faster-diffusing solutes within dense domains. In the limit of fast chemical turnover, the system always coarsens, and phase coexistence is governed by an effective free energy that explicitly depends on kinetic parameters. Beyond this limit, we develop a sharp-interface theory that predicts the onset of arrested coarsening, stationary droplet sizes, and nucleation conditions under chemical driving. Taken together, our results establish kinetic regulation as a minimal and robust mechanism to control phase coexistence and coarsening in chemically active mixtures.

The maximal positively invariant (MPI) set is obtained through a backward reachability procedure involving the iterative computation and intersection of predecessor sets under state and input constraints. However, standard static feedback synthesis may place some of the closed-loop eigenvalues at zero, leading to rank-deficient dynamics. This affects the MPI computation by inducing projections onto lower-dimensional subspaces during intermediate steps. By exploiting the Schur decomposition, we explicitly address this singular case and propose a robust algorithm that computes the MPI set in both polyhedral and constrained-zonotope representations.

In previous joint work with Tenenbaum, the truncation step $f \mapsto f_R$ in the conditional effective Erdos-Wintner theorem on the fibre $ω(n)=k$ yields, in the continuous case for real strongly additive $f$, a remainder of size $η_f(R)^{r/(r+1)}$, where $R$ is the truncation level and $r=k/\log\log x$. We prove an effective linear truncation lemma showing that, in the central window $κ\le r \le 1/κ$, this bound improves to the natural linear scale $rη_f(R)$ under an effective Sathe-Selberg-type ratio estimate for the fibre. This yields a direct effective sharpening of the truncation step in the previous joint work. The same truncation upgrade also applies to prime-set restrictions, $Ω$-fibres, and weighted fibres whenever the corresponding ratio estimate is available.

We apply one-dimensional convolutional neural networks to the Frobenius traces of elliptic curves over $\mathbb{Q}$ and evaluate and interpret their predictive capacity. In keeping with similar experiments by Kazalicki--Vlah, Bujanović--Kazalicki--Novak, and Pozdnyakov, we observe high accuracy predictions for the analytic rank across a range of conductors. We interpret the prediction using saliency curves and explore the interesting interplay between murmurations and Mestre--Nagao sums, the details of which vary with the conductor and the (predicted) rank.

Concentration gradients at the microscale play a central role in many physical, chemical, and biological systems, yet their quantitative visualization remains challenging due to the limited optical contrast associated with changes in concentration. Here, we present RIO (the Refractive Index Observer), a label-free interferometric tool for quantitative imaging of refractive index, and thus concentration fields in microfluidic systems. Implemented using a Fabry-Pérot microfluidic chip mounted on a standard optical microscope, RIO achieves a per-pixel refractive index precision on the order of $1\times 10^{-5}$ refractive index units (RIU) using a standard CMOS camera, enabling high sensitivity two-dimensional chemical imaging. We Characterize the refractive index resolution and spatiotemporal performance of the instrument and demonstrate its capabilities by measuring concentration gradients of dissolved NaCl in a co-laminar flow. RIO provides an accessible, label-free platform for quantitative studies of microscale concentration fields in systems where molecular labeling is undesirable or impractical, and enables investigations of a broad range of out-of-equilibrium phenomena, from polymerization and enzymatic reactions to cell signaling and electrochemical processes.

We consider the Bethe ansatz integrable Russian Doll (RD) model of superconductivity with time-reversal symmetry breaking, which exhibits a cyclic renormalization group. By obtaining an exact solution for the renormalization group flows, we investigate the phase structure in the one-pair sector, which includes localized, fractal, and delocalized phases. We show that the quantum number Q, arising from the Bethe ansatz equations, counts the number of cycles and parametrizes the towers of states. Using the action of the renormalization group on the eigenstates, we demonstrate that Q serves as an order parameter, providing a new mechanism for the formation of the fractal phase in the deterministic systems and an example of the interplay between fractality and cyclic RG.

Long-context question answering (QA) over lengthy documents is critical for applications such as financial analysis, legal review, and scientific research. Current approaches, such as processing entire documents via a single LLM call or retrieving relevant chunks via RAG have two drawbacks: First, as context size increases, response quality can degrade, impacting accuracy. Second, iteratively processing hundreds of input documents can incur prohibitively high costs in API calls. To improve response quality and reduce the number of iterations needed to get the desired response, users tend to add domain knowledge to their prompts. However, existing systems fail to systematically capture and use this knowledge to guide query processing. Domain knowledge is treated as prompt tokens alongside the document: the LLM may or may not follow it, there is no reduction in computational cost, and when outputs are incorrect, users must manually iterate. We present Halo, a long-context QA framework that automatically extracts domain knowledge from user prompts and applies it as executable operators across a multi-stage query execution pipeline. Halo identifies three common forms of domain knowledge - where in the document to look, what content to ignore, and how to verify the answer - and applies each at the pipeline stage where it is most effective: pruning the document before chunk selection, filtering irrelevant chunks before inference, and ranking candidate responses after generation. To handle imprecise or invalid domain knowledge, Halo includes a fallback mechanism that detects low-quality operators at runtime and selectively disables them. Our evaluation across finance, literature, and scientific datasets shows that Halo achieves up to 13% higher accuracy and 4.8x lower cost compared to baselines, and enables a lightweight open-source model to approach frontier LLM accuracy at 78x lower cost.

JWST data have enabled the abundant identification of compact broad Balmer line sources nicknamed the Little Red Dots. While they share broad lines with active galactic nuclei, they are unusually X-ray and infrared weak. We investigate the origin of the Balmer line profiles based on an empirical analysis of 18 broad H$α$-selected sources with high quality spectra at $z\approx3-7$. The H$α$ line profiles vary systematically with Balmer break strength: sources with blue UV to optical colors show a narrow core profile, redder sources with Balmer breaks a blue shifted absorption (P Cygni shape), and the reddest sources display absorption-dominated cores. All H$α$ lines have symmetric exponential wings, which are more dominant and slightly broader in red sources. Balmer absorption is present in $\sim60$ % of the sample, with H$β$ showing relatively stronger absorption. Drawing upon empirical analogies with stellar phenomena, we interpret these trends as being due to radiative processes that depend on variations in the optical depth, ionisation state and column density of a clumpy, partially ionised envelope. We unveil a correlation between the absorber velocity and Balmer break strength, with the densest absorbers inflowing and bluer sources having faster outflows. This indicates viewing angle or evolutionary effects where optically thick gas is inflowing, as suggested in models of super-Eddington accretion, and the engine can more easily drive outflows in directions with lower column densities. This new understanding of Balmer line profiles as tracing gas properties rather than dynamical broadening helps resolve tensions associated with high inferred black hole masses from standard virial calibrations, and reveals the complex gas environment around the hot central engine.

In the forthcoming years, the study of the fundamental interactions between gravitational waves (GWs) and matter will be crucial in order to understand what the new generations of GWs detectors will tell us. We present the inverse bremsstrahlung (IB) absorption of GWs as a novel approach to GWs physics that can help set constraints on different physical models. We study the absorption of GWs in scattering processes of interacting dark matter. The observation of GWs of a given frequency sets constraints on its absorption efficiency. In the case of interacting dark matter, this can translate to constraints on its mass-coupling space, or in its temperature. For this, we parametrize the absorption of GWs in DM halos and in IGM, at low and very high redshifts. We find the arising constraints to be less stringent than existing ones.

This paper analyzes the physical layer security performance of massive uplink Internet of Things (IoT) networks operating under the finite blocklength (FBL) regime. IoT devices and base stations (BS) are modeled using a stochastic geometry approach, while an eavesdropper is placed at a random location around the transmitting device. This system model captures security risks common in dense IoT deployments. Analytical expressions for the secure success probability, secrecy outage probability and secrecy throughput are derived to characterize how stochastic interference, fading and eavesdropper spatial uncertainty interact with FBL constraints in short packet uplink transmissions. Numerical results illustrate key system behavior under different network and channel conditions.

We consider database schemas consisting of a single binary relation, with key constraints and inclusion dependencies. Over this space of 20 schemas, we completely characterize when one schema is generically dominated by another schema. Generic dominance, a classical notion for measuring information capacity, expresses that every instance of a schema can be uniquely represented in the dominating schema, through application of a deterministic, generic data transformation. Our investigation is motivated both by current interest in schema design for graph databases, as well as by intrinsic scientific interest. We also consider the ternary case, but without inclusion dependencies, and discuss how the notions change in the presence of object identifiers.

Statistical offices face a familiar and intensifying dilemma: rising demand for detailed regional and domain-level estimates under budgets that are fixed or shrinking. National statistical offices (NSOs) either ignore the problem of optimal sample allocation for multiple target variables when designing a multi-purpose survey, or address it incorrectly - relying on ad hoc approaches such as computing Neyman allocations separately per variable and taking the element-wise maximum, a practice that simultaneously wastes budget and fails to guarantee precision across all domains. This paper presents a practical two-stage strategy that reframes the question: not how to allocate a given sample, but how small the sample can be made while still meeting pre-defined precision targets for all target variables across all geographic domains at once. The innovation lies not in inventing new methods, but in the novel combination of two well-established techniques applied to this cost-reduction problem: (i) multivariate constrained optimisation via Bethel allocation, which finds the globally minimum sample satisfying all precision constraints simultaneously; and (ii) Hierarchical Bayes (HB) small area modelling, which borrows strength across strata and permits a further reduction of the Bethel sample. The approach is validated using a Monte Carlo study (B = 1,000 replications) based on a synthetic labour-force population of one million individuals, where known population truth allows rigorous evaluation of precision, accuracy, and credible-interval coverage. Keywords: Bethel allocation; Hierarchical Bayes; small area estimation; sample size reduction; multivariate optimisation; labour force survey; coefficient of variation.

We study finite and semi-finite vector bundles on complex tori. We give an explicit decomposition of such bundles in terms of torsion and unipotent factors. As a consequence, we prove that the extended Nori fundamental group scheme of a complex torus decomposes as the product of its etale fundamental group scheme and its unipotent fundamental group scheme.

We calculate the heights of Stiefel--Whitney classes of the canonical vector bundle over the oriented Grassmannians $\widetilde G_{n,4}\cong SO(n)/(SO(4)\times SO(n-4))$ in the cases $n\in\{2^t-2,2^t-1,2^t,2^t+1\}$, $t\ge4$. Using some additional computations in modulo $2$ cohomology of $\widetilde G_{n,4}$ and the well-known connection between topological complexity and zero-divisor cup-length, we obtain lower bounds for topological complexity of these Grassmannians. We also extend recent results of Rusin, who computed the modulo $2$ cup-length of $\widetilde G_{n,4}$ for $n\in\{2^t-2,2^t-1,2^t\}$, to the case $n=2^t+1$, $t\ge3$.

The advent of Large Language Models (LLMs) represents a turning point in the theoretical foundations of Information Systems Engineering. Beyond their technical significance, LLMs challenge the ontological, epistemological, and semiotic assumptions that have long structured our understanding of in-formation, representation, and knowledge. This article proposes an integrative reflection on how LLMs reconfigure the relationships among language, meaning, and system design, suggesting that their emergence demands a re-examination of the conceptual foundations of contemporary information systems. Sketching on philosophical traditions from Peirce to Heidegger and Floridi, we investigate how the logic of generative models both extends and destabilises classical notions of ontology and signification. The discussion emphasises the necessity of grounding LLM-based systems in transparent, ethically coherent frameworks that respect the integrity of human-centred knowledge processes. Ultimately, the paper argues that LLMs should be understood not merely as tools for automation but as epistemic agents that reshape the philosophical and semiotic foundations of information systems engineering.

This work investigates antenna coding optimization to enhance the channel capacity of single-input single-output orthogonal frequency division multiplexing (SISO-OFDM) systems empowered by highly reconfigurable pixel antennas. We first introduce the model for pixel antenna empowered SISO-OFDM systems using a beamspace channel representation. We next formulate the problem to maximize the channel capacity through jointly optimizing antenna coding and the power allocation across subcarriers and solve it by Successive Exhaustive Boolean Optimization (SEBO) and water-filling (WF) algorithm. To reduce computational complexity, a codebook-based approach is also proposed for antenna coding optimization. Simulation results show that the channel capacity of SISO-OFDM system across all signal-to-noise-ratio (SNR) regions considered can be enhanced through leveraging pixel antennas as compared to using conventional antenna with fixed configuration. This result demonstrates the effectiveness of antenna coding technology empowered by pixel antenna in enhancing SISO-OFDM systems.

In 2023, superconductivity in La$_3$Ni$_2$O$_7$ was discovered under high pressures above approximately 14 GPa. In addition to its high transition temperature ($T_{\mathrm{c}} \simeq 80$ K), the structural resemblance to high-$T_{\mathrm{c}}$ cuprates has strongly stimulated research, soon followed by the discovery of superconductivity in La$_4$Ni$_3$O$_{10}$. These compounds belong to the Ruddlesden--Popper phases, comprising double- and triple-layered NiO$_2$ square lattices separated by LaO rock-salt slabs. Research on these systems has rapidly developed along three major directions, as in other prominent families of superconductors such as the cuprates and iron arsenides: expanding the chemical variety of compounds, enhancing $T_{\mathrm{c}}$ through elemental substitution, and elucidating the superconducting mechanism. These challenges, being closely interconnected, continue to drive the field. The clarification of the pairing mechanism encounters a particular difficulty, since the key experiments must be performed under high pressures. This situation highlights the significance of developing nickel oxides that exhibit superconductivity at much lower pressures, ideally at ambient pressure, which would in turn broaden the scope of chemical tuning and detailed physical characterization. In this context, it is timely and meaningful to summarize the present state of knowledge. Here, we emphasize sample synthesis and characterization, which are already well established and often decisive for progress in unconventional superconductors, while providing a brief overview of the currently available electronic properties.

In the context of Rydberg anti-blockade, this paper proposes a new scheme for a high-fidelity controlled-unitary gate based on non-adiabatic holonomic quantum computation. Under specific detuning and interaction conditions, the scheme achieves a suitable evolution path for non-adiabatic holonomic quantum computation through reverse engineering of pulse parameters. Numerical simulations show that the geometric gate maintains high fidelity even in the presence of spontaneous radiation and laser intensity errors. Finally,we extend our designed quantum gates to non-local gates and investigate their use in converting four-qubit entangled states. This finding indicates the potential applicability of our scheme to complex quantum information processing tasks.

Quasi-topological theories of gravity are known to resolve black-hole singularities. We investigate whether the same mechanism can remove cosmological singularities. Focusing on non-polynomial curvature quasi-topological gravities in $d=4$ dimensions, we find three generic scenarios with the correct infrared limit but without a Big-Bang singularity, for universes filled with pure radiation or other standard matter. The first scenario yields a universe emerging from a de Sitter phase, a case for which the curvature invariants remain finite but the matter density diverges, albeit only at infinite affine distance. The second one corresponds to a bouncing universe, which requires a multi-valued Lagrangian. The third possibility is an asymptotically Minkowski origin, reminiscent of an eternally loitering universe. The matter energy density for this solution is non-singular even at infinite affine distance and does not enter a super-Planckian regime, but is instead approximately constant for the past eternity.

This note introduces an extension to the definition of symphonic maps, denoted as $φ:(M,g)\longrightarrow(N,h)$, by exploring variations in the bi-energy functional associated with the pullback metric $φ^*h$ between two Riemannian manifolds.

Freudenthal algebras over a field are basically the same as Jordan algebras of degree $3$ remaining simple under all base field extensions. These algebras are intimately linked, via their automorphism groups and structure groups, to simple algebraic groups over arbitrary fields. Our main concern here will be the question of when these algebras are homogeneous in the sense that all their Jordan isotopes are isomorphic. We answer this question by presenting various necessary and sufficient conditions for homogeneity and by connecting it with the first Tits construction of cubic Jordan algebras, most notably through investigating Freudenthal division algebras over complete fields under a discrete valuation. We also study the first Tits construction in its own right by producing a local version of it, deriving a local-global principle, and by connecting it with the embeddibility of certain rank-$2$-tori into the automorphism group scheme of an Albert division algebra.

Requirements volatility is a major issue in software (SW) development, causing problems such as project delays and cost overruns. Even though there is a considerable amount of research related to requirement volatility, the majority of it is inclined toward project management aspects. The relationship between SW architecture design and requirements volatility has not been researched widely, even though changing requirements may for example lead to higher defect density during testing. An exploratory case study was conducted to study how requirements volatility affects SW architecture design. Fifteen semi-structured, thematic interviews were conducted in the case company, which provides the selection of software products for business customers and consumers. The research revealed the factors, such as requirements uncertainty and dynamic business environment, causing requirements volatility in the case company. The study identified the challenges that requirements volatility posed to SW architecture design, including scheduling and architectural technical debt. In addition, this study discusses means of mitigating the factors that cause requirements volatility and addressing the challenges posed by requirements volatility. SW architects are strongly influenced by requirement volatility. Thus understanding the factors causing requirements volatility as well as means to mitigate the challenges has high industrial relevance.

Isotopic measurements of presolar silicon carbide grains from dying stars have revealed a puzzling overabundance of $^{94}$Mo that stellar nucleosynthesis models have failed to reproduce for two decades. This discrepancy challenged our understanding of the slow neutron-capture process ($s$-process) that forges approximately half of the elements heavier than iron. The key uncertainty lies at $^{94}$Nb, a radiactive branching point where competition between neutron capture and beta decay governs the $^{94}$Mo production, yet the neutron-capture cross section had never been measured. Here we report the first experimental determination of the $^{94}$Nb(n,$γ$)$^{95}$Nb cross section important for Mo isotopic abundances. The measurement was enabled by a coordinated effort involving high-purity target preparation at Institute of Solid State and Materials Research (IFW) Dresden, radioactive sample production at the Institut Laue-Langevin (ILL) Grenoble, radiochemical characterization at Paul Scherrer Institute (PSI) Villigen, and the Time-of-Flight CERN n$\_$TOF facility using for the first time segmented total-energy detectors. Incorporation of the resulting Maxwellian-averaged cross section into fully coupled nucleosynthesis models of low-mass asymptotic giant branch (AGB) stars brings them into agreement with the presolar grain data. These results remove a major nuclear-physics input uncertainty at the $^{94}$Nb branching point and provide a firmer foundation for understanding the origin of $^{94}$Mo in the solar system.