Earth and Planetary Astrophysics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Tuesday 8 September
With microarcsecond precision, Gaia has made the astrometric discovery of substellar companions feasible. However, follow-up observations are needed to check on orbital solutions and rule out astrophysical false positives. We validate and characterise a sample of 20 Gaia Data Release 3 (DR3) astrometric candidates for substellar companions, with the dual goal of identifying false positives and deriving robust physical parameters for genuine companions. We performed high-resolution spectroscopy using FIES, NIRPS, and NEID at the Nordic Optical Telescope, the ESO 3.6m telescope, and the WIYN Telescope, respectively. Double-lined spectroscopic binaries were identified as false positives, and surviving systems were characterised through joint Gaia astrometry and radial velocity modelling. Thirteen of the candidates proved to be near-twin binary stars for which the astrometric motion of the center of light is small enough to mimic that of a single star in response to a substellar companion. We confirm that six companions are brown dwarfs with masses in the range ${\sim}26$–$68\,M_\mathrm{Jup}$, two of them for the first time. In contrast to earlier studies of lower-mass Gaia candidates, we find that the astrometric orbital solutions are generally reliable to within the stated uncertainties.
Software Engineering
The Unreliable Progress Bar: Can LLM Agents Reliably Report Task Progress Throughout Execution?
Recent large language models can emit task-progress signals that agent frameworks use to decide whether a task should continue or stop, yet whether a model can reliably report its task progress at every stage of a task, and where and how its reports fail, has not been studied systematically. We evaluate this ability on the public benchmark $\tau^2$-bench and on StageIF, a controlled testbed in which reporting checkpoints are placed across the task's lifecycle. Both settings require reports at multiple task stages. We find that reporting reliability depends on the stage a task has reached, and that almost every deployed model we test is reliable at some stages and unreliable at others. Where reporting breaks down is not the same everywhere. Most deployed models lose accuracy once work is under way and recover once the task is done. The newest generation closes that mid-task drop and instead grows conservative at the finish line. Our study exposes a capability gap in task-progress reporting and provides an evaluation protocol that spans the whole course of task execution for this ability on which agent operation depends. The findings indicate that agent frameworks should not control task flow on the strength of the model's state reports alone.
Read the full abstract
Combinatorics
A Fano framework for binary delta-matroids
Dunshee and Ellingham recently showed that seven natural properties of a cellularly embedded graph form a Fano-plane framework. We establish an analogous framework for binary delta-matroids. For a binary delta-matroid $D$ and $\tau$, let $Z_3(D,\tau)$ denote its associated binary tight $3$-matroid. The six outer points are represented by evenness or bipartiteness of $D$ and its global vertex-flip transforms. For the seventh point, we call $D$ $Z_3$-bipartite when every circuit of $Z_3(D,\tau)$ has even cardinality. We show that the satisfied properties are precisely the nonzero vectors of a subspace of $\Ftwo^3$. For ribbon-graphic delta-matroids, $Z_3$-bipartiteness is equivalent to bipartiteness of the medial graph, so the construction recovers the Fano-plane framework for embedded graphs.
Read the full abstract
Human-Computer Interaction
Rescuing Performance from the Demo: Co-Designing Drum Gesture Mappings with a Percussionist
Augmenting instruments with sensors and neural network mappings is a well-explored digital musical instrument design approach. While augmentations can create new expressive opportunities, they also exert aesthetic influence and can constrain musicians' gestural language, which, if left unchecked, can lead to technological capture. To examine this, we conducted a study with a professional percussionist, co-developing a gesture mapping toolkit and recording a ten-track album. Drawing on the concept of productive dissonance, our study aimed to hold the musician's aesthetic in tension with technological constraints. This, along with a practice-based reflective approach, supported the development of a continuous gesture recognition method for percussive mapping and surfaced insights into the design process. We identify knowing-when as a form of tacit knowledge that supported productive dissonance, and raise an open question: absent a musician's broader social context, how do we know whether a technology's influence is genuinely supporting their practice?
Read the full abstract
Analysis of PDEs
On the Invertibility of the Potential-Density Mapping for the Vlasov-Poisson System in Analytic Spaces
We develop the mathematical analysis of the Time-Dependent Density Functional Theory introduced by Manfredi [J. Plasma Phys., 86.2 (2020), 825860201] in the setting of Vlasov equations with a time-dependent exterior potential. In this context, we establish the invertibility of the potential-density mapping for the Vlasov-Poisson system on $\mathbb{T}^d \times \mathbb{R}^d$ when the exterior potential is (real-)analytic in the space variable. We also recover a Runge-Gross type Theorem for the Vlasov-Poisson equation for smooth (real-)analytic in time exterior potential. The surjectivity of the potential-density mapping is an inverse problem for the Vlasov-Poisson equation: we find the electric field that generates a given density profile on a given time interval.
Read the full abstract
Computation and Language
Limitations of Automated Simulatability: LLM Simulators Can Bypass Explanations
Simulatability is an evaluation protocol for explanations that quantifies their usefulness by how well they help a user predict a task model's outputs. Since human evaluation is costly, automated simulatability replaces human explainees with LLM simulators, as proposed in ConSim (Poché et al., 2025) for large-scale experiments. We qualitatively replicate and extend ConSim's ranking of explanation methods across the tested datasets, explanation families, and simulator LLMs, and identify two limitations. First, when class names are meaningful, simulators can obtain high simulatability by solving the classification task directly, without relying on the explanations. Second, class anonymization can reward explanations for leaking the hidden label mapping, a limitation we expose with a new classes-as-concepts baseline. These results are consistent with a shortcut hypothesis: in the tested settings, simulator predictions mainly rely on task priors, while explanations produce small changes. We derive recommendations for more robust automated simulatability evaluations.
Read the full abstract
High Energy Physics (Theory)
Gravitational decoherence from soft-graviton $S$-matrix
We revisit gravitational decoherence of matter superpositions using the leading soft sector of the graviton $S$-matrix. Starting from Weinberg's soft theorem and the overlap of coherent radiation states, we derive, for branch configurations with the same total four-momentum, a gauge-invariant, nonnegative leading-soft decoherence exponent and evaluate the polarization sum and angular integrals exactly at this order. In the vacuum, coherence between branches with distinguishable soft radiation vanishes as the infrared cutoff is removed. The leading nonrelativistic term is purely quadrupolar as a consequence of energy–momentum conservation. We extend the analysis to single- and two-mode squeezed graviton states, including the two-mode case motivated by inflation. Squeezing can enhance or suppress decoherence depending on the relative phase, while phase averaging gives an enhancement governed by the graviton occupation. For a phase-averaged relic background with an approximately flat energy-density spectrum over the relevant soft frequency band, the sharp-cutoff identification of the infrared scale with the inverse observation time gives a contribution that grows with the gravitational-wave energy density and with the fourth power of the observation time. These results connect soft-graviton scattering, infrared quantum information, and decoherence in nonvacuum gravitational backgrounds.
Read the full abstract
Robotics
Estimating Semantic Ambiguity via Gaussian Context Distributions for VLM-Driven Traversability Analysis
Autonomous navigation in unstructured environments requires robust scene understanding, yet Vision-Language Models (VLMs) often suffer from semantic ambiguity, where conflicting predictions can lead to dangerous failures. To address this, we present a novel pipeline for vision-based traversability estimation that explicitly models contextual uncertainty. Our approach utilizes Conceptual Anchoring to ground open-vocabulary VLM predictions onto a continuous physical traversability scale. By formulating the model's responses as a Gaussian Context Distribution (GCD), we derive both a dense traversability map and a dense uncertainty map based on the statistical properties of the distribution. Experimental validation on the real-world GOOSE dataset demonstrates that our proposed uncertainty metric effectively correlates with sources of ambiguity, such as visual artifacts and mixed terrain overlap. The method exhibits competitive performance while offering the distinct advantage of providing statistical uncertainty estimates to address semantic ambiguity, enabling safer and more reliable autonomous behavior in complex outdoor settings.
Read the full abstract
Machine Learning
Leveraging Cardiac Imaging to Improve ECG-Based Detection of Chagas Disease in Resource-Constrained Settings
Chagas disease is a major cause of cardiomyopathy in Latin America. Cardiac magnetic resonance (CMR) imaging can characterize its structural abnormalities, but scanners and expert readers remain scarce in endemic regions. Electrocardiography (ECG) is inexpensive and widely available, yet structural disease must be inferred indirectly from electrical signals. We propose to transfer CMR-derived structural knowledge to ECG through contrastive pre-training. Using 63,193 paired ECG-CMR examinations from the UK Biobank, we align an ECG encoder with a clinically grounded CMR embedding space using an asymmetric InfoNCE objective. Despite seeing no Chagas cases during pre-training, the resulting representation improves ECG-based Chagas detection. Across CODE-15% and SaMi-Trop, a frozen linear probe achieves an AUROC of 0.851 and sensitivity at the top 5% of predicted risk (Top5%-TPR) of 0.427 in five-fold cross-validation, compared with 0.827 and 0.377 for an unaligned ECG-FM baseline. On the PhysioNet/CinC 2025 Challenge test set, our model obtains the highest AUROC on SaMi-Trop-3 and the best ELSA-Brasil challenge score among the three top-performing methods, indicating that imaging-supervised ECG representations can generalize to populations and settings beyond the pre-training distribution.
Read the full abstract
Machine Learning
AlphaRJM: Reward-Jump Memory for Stochastic Return-Guided Alpha Discovery
Formulaic alpha discovery is a pool-dependent symbolic search problem in which informative feedback is observed primarily when a complete expression is evaluated. This delayed feedback creates two coupled difficulties: the retained alpha pool does not preserve the full history of realized evaluation feedback, and the value of an intermediate construction action is uncertain because its consequence depends on the formula eventually completed. We introduce AlphaRJM, which addresses these difficulties through Reward-Jump Memory, an event-driven latent state that remains fixed during token construction and updates only at terminal evaluation events using the realized pool reward and evaluation outcome, and an action-conditioned SDE return critic that represents future discounted discovery returns with stochastic particles. The particles guide action selection through their mean and uncertainty and are learned using a distributional Bellman objective combining energy-distance matching, mean calibration, and jump regularization. Empirically, AlphaRJM delivers strong and stable gains across multiple equity universes, forecasting horizons, and random seeds, while ablations confirm the complementary roles of persistent evaluation history, stochastic return modeling, and distributional supervision.
Read the full abstract
Strongly Correlated Electrons
Magnetic phases of Kondo lattice materials Ce$_5$RhGe$_2$ and Ce$_5$IrGe$_2$
Single crystals of Ce$_5$RhGe$_2$ and Ce$_5$IrGe$_2$ have been systematically investigated by electrical resistivity, specific heat, and magnetization measurements. Together with Ce$_5$CoGe$_2$, all three compounds crystallize in the orthorhombic Pnma structure, with the lattice parameters increasing monotonically from Co to Rh to Ir, consistent with the effect of negative chemical pressure. Magnetization measurements along the three principal crystallographic axes identify the a axis as the easy magnetization direction throughout the series. Ce$_5$RhGe$_2$ exhibits ferromagnetic ordering with a Curie temperature of approximately 11.5 K and shows magnetic behavior closely resembling that of Ce$_5$CoGe$_2$. In contrast, Ce$_5$IrGe$_2$ undergoes two successive magnetic transitions at $T_{\rm M1}=12.7$ K and $T_{\rm M2}=11.8$ K, and there are multiple metamagnetic transitions under magnetic fields, giving rise to magnetization plateaus at fractions of the saturation magnetization $M_{\rm s}$ of approximately $M_{\rm s}/5$ and $M_{\rm s}/3$. The low-field metamagnetic transition along the easy axis shifts to lower field with decreasing temperature, and eventually a pronounced hysteresis loop is observed about zero-field, establishing that Ce$_5$IrGe$_2$ exhibits a ferrimagnetic ground state at the lowest measured temperatures.
Read the full abstract
Differential Geometry
Mass-Capacity Rigidity for Asymptotically Flat Manifolds with Arbitrary Ends
We characterize equality in the mass-capacity inequality for asymptotically flat manifolds with arbitrary ends in every dimension \(n\ge 3\). If \(u\) is the capacity minimizer, then the equality forces \((M,u^{4/(n-2)}g)\) to be isometric to \((\mathbb R^n\setminus S, g_{\mathrm{Euc}})\) for a compact set \(S\) satisfying \(\operatorname{dim}_{\mathcal H}S\le (n-2)/2\). A key ingredient is that conformal changes by positive harmonic functions transform nonnegative Ricci curvature into nonnegative Bakry–Émery Ricci curvature of effective dimension \(4-n\).
Read the full abstract
Programming Languages
Erased Postulates, Identity Types and Quotients
This text is concerned with the question of whether, in type theory with erasure annotations, one can postulate that some type is inhabited and still have a guarantee that a program will not get stuck. Previous work has provided such guarantees for consistent erased postulates, i.e. postulates that are restricted to be used in erased contexts. Here those guarantees are extended to type theory with identity types. Similar ideas provide a simple way to support quotient types: it is shown that one can let things like "the equivalence classes for two related values are equal" be erased postulates and have an eliminator that only computes for the equivalence class constructor, and still get a guarantee that programs will compute correctly. Another question is whether programs compute correctly if one is allowed to transport (cast) using erased identity proofs. It is shown that this is safe in the absence of quotients and postulates, and in the presence of quotients and erased postulates that can be implemented using equality reflection. However, unrestricted transports of this kind are not compatible with erased, postulated univalence. For that reason the text includes a study of the function []-cong, which encapsulates a limited form of transport for erased identity proofs. The text is accompanied by machine-checked Agda proofs.
Read the full abstract
Statistical Methodology
Instability in Patient Clustering: A Multiverse Analysis of Unsupervised Clustering in the CENTER-TBI cohort
Understanding patient heterogeneity is key to improving prognostic modeling in traumatic brain injury (TBI). Unsupervised clustering is widely used to explore patterns in patient characteristics that may define subgroups. However, it involves a multitude of decisions, including the choice of algorithm, the distance metric, and the method used to determine the "optimal" number of clusters. The aim of this study is to investigate how these choices influence the resulting clustering solution. We analyzed data from 4,509 patients enrolled in the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) study. K-medoids, agglomerative, and spectral clustering were applied in a complete 3 X 2 X 2 factorial design, in combination with Euclidean or Gower's distances, and silhouette score or gap statistic to choose the number of clusters. We investigated the agreement of clustering solutions with UpSet Plots and stability with the (adjusted) Rand index. Comparisons were made both across approaches using the original dataset and within approaches using bootstrap resampling. Clustering results varied substantially depending on the analysis choices. The number of suggested clusters varied widely, from one to twenty-five. Adjusted Rand indices confirmed low concordance between methods. Moreover, none of the clustering solutions demonstrated discriminatory performance comparable to a supervised logistic regression model in classifying patient recovery illustrating the limited usefulness of clustering for this purpose. The high instability in clustering results compromises interpretability and underscores that such solutions should not be blindly interpreted as underlying structure.
Read the full abstract
Computation and Language
Which Forms of Caregiver Feedback Support Grammar Learning? A Reinforcement-Learning Study of Child-Like Language Models
Social interaction is central to children's language learning, but the effects of different forms of caregiver feedback are difficult to isolate in naturalistic data. We use child-like language models as controlled learners to test which forms of feedback support grammatical development. Small GPT-2-style models are pretrained on child-directed language from CHILDES, then fine-tuned with reinforcement learning using reward models trained to capture four feedback types: communicative feedback, structural alignment, semantic contingency, and affective feedback. Reward fine-tuning yields limited gains on minimal-pair evaluations, but clearer effects in free generation. Structural alignment produces the strongest improvements in grammaticality, providing a novel, plausible mechanistic account of how this feedback can support grammar learning. Communicative feedback yields more moderate gains. In contrast, semantic contingency and affective feedback do not improve grammaticality, although further analyses suggest that they may support other aspects of language learning beyond grammar. These results suggest that different forms of caregiver feedback make complementary contributions to language learning.
Read the full abstract
High Energy Physics (Phenomenology)
Free $\mu^{+}\mu^{-}$ and bound-free $e^{+}e^{-}$ pair production in relativistic heavy-ion collisions
In our previous studies, we calculated the cross-section for the production of free $\mu^{+}\mu^{-}$ pairs using a probability function dependent on the impact parameter, and also, we computed bound-free $e^{+}e^{-}$ pair production cross section. In this study, unlike the others, we obtained the cross section for the production of bound-free $e^{+}e^{-}$ pairs using the probability function dependent on the impact parameter. Here also, we simultaneously calculated the cross sections for producing free $\mu^{+}\mu^{-}$ with bound-free $e^{+}e^{-}$ pairs in relativistic $Pb+Pb$ collisions at the LHC with Lorentz factors at $\gamma=1500$ and $\gamma=3000$ reported in the literature. Our calculation is based on second-order perturbation theory, implemented by using Monte Carlo integration techniques. We present impact parameter dependent pair production probabilities for free $\mu^{+}\mu^{-}$ and bound-free $e^{+}e^{-}$ pair production processes. Notably, the bound-free $e^{+}e^{-}$ pair production was evaluated by using a formalism adapted from that of free $\mu^{+}\mu^{-}$ pair production. Our results obtained here for the free $\mu^{+}\mu^{-}$ with bound-free $e^{+}e^{-}$ pair production cross sections are compared with those from other theoretical methods available in the literature. This study aims to provide predictions that can be tested against future experimental data, thereby offering a valuable check for quantum electrodynamics (QED) in strong electromagnetic fields.
Read the full abstract
Computation and Language
Do New Attention Mechanisms Actually Fix Attention Sinks at Million-Token Context?
Long context language models now advertise windows of one million tokens, but two habits limit how much of that window is used. Attention heads with nothing useful to read still spend their budget on the first token, which is called the attention sink, and where a fact sits in the context changes whether the model finds it. Gated attention cut first token attention from 46.7 percent to 4.8 percent at NeurIPS 2025, and Kimi K3 pairs that idea with Kimi Delta Attention and Attention Residuals behind a one million token window, eight times past the range where these diagnostics have been reported. This paper asks whether the fix survives that jump. We build SinkProbe, a suite that measures sink mass, massive activation, position resolved recall and the recency gap, and apply it to four small models that differ only in how they mix tokens and depth. Three results follow. The training objective produces the sink, not the architecture. Gating did not reproduce its published effect at our scale. Sink mass, activations and position bias moved independently. Code, data and the measurement protocol are released at https://github.com/sararizwan7/Attention-Mechanisms-in-1M-Context-Window
Read the full abstract
Cosmology and Nongalactic Astrophysics
Harnessing stellar kinematics to constrain dark energy with the double-source-plane gravitational lens SDSS J0946+1006
SDSS J0946+1006 is an attractive target for measuring cosmological parameters through lens modelling. It is the best-studied galaxy-scale strong gravitational lens with multiple sources whose redshift separations are well-suited to constraining the dark energy equation of state. However, multi-plane lens models with free cosmological parameters risk a multi-plane mass-sheet degeneracy, although this can be lifted by a non-lensing deflector density profile tracer. We simultaneously reconstruct near-infrared and near-ultraviolet HST imaging whilst including a velocity dispersion measurement from VLT-MUSE to constrain the foreground deflector. Imaging is reconstructed faithfully regardless of whether the inferred kinematics are realistic, though we find the kinematic constraint essential for shifting the preferred cosmology into a region not in significant tension with other dark energy probes. Deflector density profile perturbations, via substructure and multipolar halo shape deformations, have only a modest effect on inferred cosmology. Combining our fiducial model with Planck CMB data yields $w=-1.01^{+0.08}_{-0.13}$; or, combined with Pantheon SNe Ia, $w=-0.99^{+0.13}_{-0.15}$. We further show that this system proves a remarkably valuable complementary probe in the $w_{0}$-$w_{a}$ plane of an evolving dark energy model, and that the DESI BAO tension with $\Lambda$CDM seen when combined with other datasets is not reproduced when combined with this lens, yielding $(w_{0}, w_{a})=(-0.87^{+0.10}_{-0.11}, -0.22^{+0.28}_{-0.25})$. The system's third source, visible with MUSE, only weakly constrains $w$CDM but may strengthen $w_{0}w_{a}$CDM constraints, though further mass-model complexity along its line of sight is required. Overall, kinematics-informed multi-plane lens modelling is a robust route to competitive dark energy constraints, even with a single system.
Read the full abstract
Artificial Intelligence
AgentGrad: Intervention-guided Prompt Optimization for Multi-Agent Systems
Large language model (LLM)-based multi-agent systems (MAS) achieve strong performance by employing specialized multiple agents, yet their performance depends on the prompt design of each agent. For MAS prompt optimization, textual gradient methods that guide prompt updates using natural-language feedback have emerged as a leading paradigm. In this paper, we identify limitations in two stages of existing textual gradient approaches: gradient extraction and gradient aggregation. In gradient extraction, previous works select a prompt to update without verifying whether modifying it resolves the failure, and derive gradients without agent-level supervision over the corresponding agent's intermediate output. In gradient aggregation, individual gradients are randomly grouped and concatenated, often mixing unrelated failure modes and producing prompts that fail to generalize. To address these limitations, we propose AgentGrad, a prompt optimization framework for multi-agent systems based on sequential intervention and semantic textual gradient abstraction. For each failure, sequential intervention injects a hint into one agent at a time to identify the target agent whose modification resolves the failure. The modified output of the target agent then serves as agent-level supervision for extracting a fine-grained gradient. Semantic textual gradient abstraction clusters semantically similar gradients to prevent mixing unrelated failure modes, and abstracts each cluster into a generalized gradient that captures the shared corrective pattern. Experimental results show that AgentGrad achieves state-of-the-art performance across five MAS benchmarks while reducing wall-clock optimization time by 2.5 times and optimization cost by 21.8% on average compared to the next-best baselines. Our code is available at https://github.com/mlvlab/AgentGrad.
Read the full abstract
Formal Languages and Automata
Solution to Bucher's density problem for context-free languages
In 1980 Bucher asked whether, given context-free languages $L\subseteq U$ with $U\setminus L$ infinite, there must be a context-free language $K$ between them for which both $K\setminus L$ and $U\setminus K$ are infinite. We give a negative answer. We first construct an infinite language $D$ with context-free complement such that, for every regular language $R$, either $D\cap R$ or $D\setminus R$ is finite. The words of $D$ encode computations of factorials; repetition of letters ensures that each finite automaton either accepts all but finitely many words of $D$ or rejects all but finitely many words of $D$, while a one-counter automaton recognizes errors in the encodings. We then construct $L$ and $U$ from the complement of $D$. A grammar argument shows that any context-free intermediate language $K$ would divide $D$ in the same way as some regular language. This proves the required impossibility. Both $L$ and $U$ can be taken over a binary alphabet.
Read the full abstract
Computer Vision
Effects of model architecture and learning strategies on deep learning-based recognition of activated sludge microscopic images and comparison with quantitative image analysis
Microscopic image analysis has long been recognized as a promising approach for monitoring activated sludge. In recent years, deep learning-based image analysis has been increasingly adopted in this field because of its high performance. However, previous studies on microscopic image analysis of activated sludge have rarely explored transformer-based models or self-supervised foundation models and have instead relied on CNNs and supervised ImageNet pretraining. In addition, previous studies often downsampled image sizes, but the effects of downsampling have not been sufficiently investigated, and the relationship between downsampling strategies and image analysis performance remains unclear. Furthermore, no study has quantitatively compared deep learning performance with quantitative image analysis (QIA), which was widely used before the emergence of deep learning. In this study, to examine how model architecture and learning strategies affect performance in microscopic image analysis of activated sludge and to quantitatively determine whether deep learning outperforms QIA, we prepared three types of activated sludge samples, classified their microscopic images, and evaluated classification accuracy. Our results showed that transformer-based architectures and alternative pretraining methods were effective in terms of classification accuracy. Our downsampling analysis showed that using overly small images reduced accuracy, but increasing image size beyond a certain point did not improve it further. In addition, the analysis indicated that, to achieve high classification accuracy, maintaining the field of view was a more effective downsampling strategy than maintaining resolution. Finally, our comparison between deep learning and QIA showed that deep learning outperformed QIA in terms of accuracy.
Read the full abstract
Superconductivity
Fabrication And Characterization Of High-Quality Nb/Al-AlO$_x$/Nb Cross-Type Josephson Tunnel Junctions Utilising CMP-based Planarisation Techniques
Josephson tunnel junctions (JJs) are the fundamental building blocks of today's most advanced superconducting electronic components, such as qubits and superconducting quantum interference devices (SQUIDs). Given the ongoing demand for scalability of these devices on the wafer-scale, it is crucial to control the fabrication process as precisely as possible to ensure uniform quality and reproducibility. Today, window-type JJs are often used in the production of DC SQUIDs and while they are reliable and reproducible, they suffer from alignment inaccuracies caused by photolithography and unavoidable parasitic capacitances, thus limiting the energy sensitivity of DC SQUIDs. These problems can be circumvented by producing JJs with cross-type geometry, which allows for smaller junction areas and eliminates parasitic capacitances. Here we discuss the fabrication of Nb/Al-AlOx/Nb-based cross-type JJs, where the trilayer is embedded in sputter-deposited SiO$_2$ to allow for planarisation of the structures and to ensure the reliable insulation of the sidewalls of the bottom electrode. Instead of lift-off processes that require a lot of time and potentially leave unwanted wings along the edges of microstructures behind that might compromise subsequent layers, we use chemical-mechanical polishing (CMP) for the removal of excess SiO$_2$. This is not only much faster, but produces smooth and uniform surfaces, which in turn effectively improves the reliability of our JJ production process resulting in a high yield of over $90\,\%$ on wafer-scale. In addition, we discuss the influence of fabrication details on the quality and electrical properties of our JJs with different junction areas down to $1\,\mathrm{\mu m} \times 1\,\mathrm{\mu m}$ distributed across the wafer and extract the junction specific quality parameters from their IV-characteristics and their Fraunhofer patterns.
Read the full abstract
Classical Analysis and ODEs
Intersections and Minkowski Sums of Four-Corner Cantor Dusts with the Unit Circle
For $0<\lambda<1/2$, let $K_{\lambda}$ be the attractor of the iterated function system $\{\lambda x, \lambda x+1-\lambda\}$, and put $C_\lambda=K_{\lambda}\times K_{\lambda}$. We study the intersection \[ E_\lambda=C_\lambda\cap S^1 \] and the Minkowski sum \[ A_{\lambda}=C_\lambda+S^1, \] where $S^1$ is the unit circle. For the intersection problem, we prove that $E_\lambda$ has the cardinality of the continuum for $(\sqrt{3}-1)/2<\lambda<1/2$, and $\dim_{\rm H} E_\lambda>0$ for $\sqrt{2}-1<\lambda<1/2$. We also establish quantitative lower bounds for $\dim_{\rm H} E_\lambda$ for $\lambda$ near $1/2$; in particular, $\dim_{\rm H} E_\lambda$ approaches $1$ as $\lambda\uparrow1/2$. For the Minkowski sum problem, we prove that $A_{\lambda}$ has nonempty interior throughout the previously open range $1/4<\lambda<1/3$, answering a question of Simon and Taylor. Together with earlier results of Simon and Taylor, our theorem yields the complete classification: $A_{\lambda}$ has nonempty interior in $\mathbb{R}^2$ if and only if $1/4<\lambda<1/2$. More generally, we prove that $C_\lambda+\Gamma$ has nonempty interior for every regular $C^1$ closed curve $\Gamma$ whenever $1/4<\lambda<1/2$.
Read the full abstract
Optimization and Control
Structured Positive-Definite Optimal Control Synthesis of Closed-Loop Recommendation Systems over Social Networks
We study the design of feedback recommendation policies for networked multi-topic opinion dynamics. The design of recommendations is formulated as an infinite-horizon linear-quadratic optimal control problem that favors suggestions aligned with each agent's current opinion, thereby using opinion alignment as a proxy for engagement. At the same time, the performance index penalizes polarization, deviation from the uncontrolled equilibrium of the opinion dynamics, and from the current agents' opinion, recommendation effort, and favors coherence among neighbors' recommendations. An agent-wise representation exposes the interconnection structure of the social network and enables structured controller and certificate synthesis. Under the assumption that the stage cost of the performance index is strictly positive definite, the affine optimal control problem separates into a strictly convex steady-state optimization, which determines the optimal equilibrium and the affine controller offset, and an LQR problem, which yields the static state-feedback gain. The resulting centralized Riccati controller provides a performance benchmark, while structured gains are obtained from dissipativity-based and $\mathcal H_2$-based LMI surrogate formulations. Possibly overlapping certificate clusters yield scalable local sufficient conditions for the structured gain design, while a separate steady-state cluster decomposition provides an exact consensus reformulation of the steady-state optimization. Numerical results on a social network illustrate the closed-loop behavior and the trade-off between performance and controller locality.
Read the full abstract
Artificial Intelligence
BIO-MEMART: Biometric-Aware KV Cache Memory for Multi-User LLM Agents
KV cache is evolving from a serving optimization into an external memory substrate for long-term LLM agents. In a shared multi-user deployment, however, reusable KV blocks introduce a missing access-control question: semantic relevance alone cannot determine whether a memory block is authorized for the current physical user. We propose Bio-MemArt, a biometric-aware KV-cache memory framework for multi-user LLM agents. Bio-MemArt attaches a normalized biometric template to each stored KV memory block, filters the shared memory pool with the current user's biometric probe, and then runs the original MemArt retrieval and KV reuse pipeline only inside the authorized candidate pool. This design preserves latent-space retrieval, direct cache reuse, and decoupled position encoding while adding physical-user access control to shared KV memory. We evaluate Bio-MemArt under Owner and Non-owner query conditions on long-term dialogue QA with face and palmprint benchmarks. Across face benchmarks, the average owner and non-owner biometric success rates are 95.71% and 0.86%; across palmprint benchmarks, they are 97.60% and 2.00%. In the efficiency study, average prefill tokens drop from 18,781.96 under full-context prompting to 28.57 with Bio-MemArt, showing that biometric gating preserves the low-token operating regime of KV-cache memory.
Read the full abstract
Algebraic Geometry
On the $a$-number of Fermat type function fields and some of their subfields
The $a$-number of a function field $F$ is the dimension of the kernel of the Cartier operator acting on the holomorphic differentials of $F$. We compute the action of the Cartier operator on the holomorphic differentials on the Hermitian function field $\mathcal{H}$, function fields of Fermat type $\mathcal{F}_{n,m}$ and the Hurwitz function field $\mathcal{H}_n$. From this, we obtain various results about the $a$-number of these function fields, including a generalisation of the results on the Fermat function field from Kodama and Washio.
Read the full abstract
Machine Learning (Statistics)
Non-Adaptive 1-Bit Mean Estimation: Minimax Rates and the Sample-Interval Tradeoff
We study distributed one-dimensional mean estimation under a 1-bit communication constraint. Each agent observes one sample, drawn independently from an unknown distribution, and returns a single bit in response to a query $Q: \mathbb{R}\to\{0,1\}$ chosen by a central learner. The distribution has mean in $[-\lambda,\lambda]$ and $k$-th central moment at most $\sigma^k$, for a fixed $k>1$. The order-optimal two-stage protocol of Lau and Scarlett uses responses from the first batch to choose the second-batch queries, motivating the question of whether this single round of interaction is necessary. We answer this negatively: for every $k>1$, a non-adaptive protocol attains the adaptive 1-bit minimax rate (and concurrent works reached the same conclusion via different strategies). We further determine the minimax sample complexity among non-adaptive 1-bit estimators when every one-set $Q^{-1}(1)$ is restricted to a union of at most $s$ intervals. Relative to unrestricted non-adaptive 1-bit querying, this constraint adds a term of order $(\lambda\sigma/(s\varepsilon^2))\log(1/\delta)$, giving the full tradeoff between sample complexity and interval complexity to within $k$-dependent constant factors. As a corollary, we identify, order-wise, the minimum interval budget needed to retain the unrestricted 1-bit minimax sample rate.
Read the full abstract
Software Engineering
PLC-Bin2Src: Retrieving Corresponding Structured Text Source Files for PLC Binaries
PLC developers reuse existing components and third-party libraries to reduce the cost of developing new applications, but this reuse can also introduce security risks. When components are distributed only in binary form, provenance tracing, vulnerability assessment, and security auditing become more difficult. To support these analyses, binary2source matching can retrieve the corresponding Structured Text (ST) source file for a PLC binary from a candidate source repository. However, due to the properties of PLC software, this task faces three challenges: cross-platform compilation heterogeneity, a binary-source representation gap, and inadequate representations of PLC semantics. Therefore, we present PLC-Bin2Src, a cross-platform binary2source matching framework that retrieves corresponding ST source files for binaries produced by CODESYS, GEB, OpenPLC v2, and OpenPLC v3. First, PLC-Bin2Src uses platform-aware binary frontends to recover user control logic across toolchains. Then, ST-to-C conversion and shared normalization are used to reduce syntactic differences between source- and binary-side representations. Next, control-data flow graphs (CDFGs) and function call graphs (FCGs) are constructed to capture program semantics, while recovered symbols provide complementary identity evidence. Finally, PLC-Bin2Src compares these representations and combines their similarity scores to rank source candidates. We evaluate PLC-Bin2Src on PLC-BEAD. The results show that PLC-Bin2Src achieves 95.89% Recall@1, 99.66% Recall@5, and an MRR of 0.9769 across four PLC platforms.
Read the full abstract
Combinatorics
Matchings and shape-Wilf-Equivalence of sets of patterns of length three I: Triples
Permutation pattern avoidance on Ferrers boards has become a central topic in enumerative combinatorics with important connections to matchings, set partitions, and other combinatorial structures as it allows one to build families of Wilf-equivalent patterns. While shape-Wilf-equivalence classes have been completely determined for individual patterns and pairs of patterns of length three, the corresponding classification for larger pattern sets has remained open. In this paper, we provide a complete classification of the shape-Wilf-equivalence classes of triples of patterns of length three. Our proofs use a bijective encoding of pattern avoiding transversals to establish all equivalence classes. As an application, we enumerate matchings avoiding triples of patterns of length three for all, extending previous results of Bloom and Elizalde. These enumerative results identify additional families of combinatorial objects counted by the Fuss-Catalan numbers and by other integer sequences appearing in the OEIS.
Read the full abstract
Machine Learning
Certified Topological Interaction in Neural Representations: Exact Tests and the Statistic They Require
Class disentanglement–the separation of a representation's class-conditional point clouds along depth and over training–is measured by descriptive curves: the sentence such a study wants to write, layer l+1 is more disentangled than layer l, is an eyeball judgement with no null. We supply the inferential layer for a topological measurement of class overlap, the Intersection Euler Characteristic Profile: the Euler characteristic of the overlap of the clouds' ball unions as a function of scale, from one Alpha-complex sweep with no boundary-matrix reduction. Every number carries a test–exact permutation tests in both directions, a guarded separation certificate the invariant requires, and a paired sign-flip test for comparative claims. Building that test taught a lesson outliving this invariant: its statistic must be scale-free. On the raw profile mass, which has units of feature length, 12,375 paired tests return 5,633 significant steps of which every one at the first epoch points the wrong way, certifying feature-norm dynamics as disentanglement; the dimensionless statistic returns 2,707, with 2,026 decreases. Across 111 networks and 52,650 measurements, disentanglement is depth-graded and early, and interaction quotients rank class pairs by confusability (rho=0.83), on par with cheap separability statistics. In a 96-model factorial, augmentation is the one training choice that separates classes relative to chance; weight decay compresses the overlap without separating. Only a k-fold statistic can pose the structural question: the joint entanglement of a class triple sits below its strongest pair in 97% of triple-layer cells and 99.5% of deep cells, at median ratios far below a measured null floor, in vision encoders and frozen language models–a regularity, not a law. The unnormalized mass predicts test accuracy (R^2=0.94), the quotient does not, and neither beats a linear probe.