Papers

Passive-wheeled terrestrial-aerial bimodal vehicles (TABVs) combine aerial mobility with energy-efficient ground locomotion. However, reliable air-ground mode switching under limited onboard perception and robust ground trajectory tracking across diverse terrains remain challenging when targeting real-world applications. In this work, we propose a learning-based air-ground motion control framework for passive-wheeled TABVs: 1) a learned mode selector for autonomous air-ground motion mode switching. The selector uses historical single-point time-of-flight (ToF) measurements and robot states together with future reference information to determine the active locomotion mode. 2) a reinforcement learning control policy for trajectory tracking. The policy combines proprioceptive observations with future reference information to anticipate trajectory changes. For ground locomotion, multi-terrain training and dynamics randomization enable robust tracking across different terrains. Simulation and real-world experiments demonstrate reliable air-ground switching under limited perception and accurate ground tracking across diverse terrain conditions. The learned selector outperforms a rule-based mode selector in challenging transitions, while the ground controller achieves lower position RMSE than PID across all tested conditions and maintains decent tracking where NMPC fails. With these capabilities integrated, the system tracks a 101m air-ground trajectory through multiple autonomous mode transitions with a position RMSE of 0.08m.

The Ramsey number $r_k(s,n)$ is the smallest integer $N$ such that every $N$-vertex $k$-graph contains either a copy of $K_s^{(k)}$ or an independent set of size $n$. Erdős and Hajnal conjectured that for every fixed $s>k\ge 4$, one has $r_k(s,n)\ge \operatorname{twr}_{k-1}(\Omega(n))$. This conjecture was independently verified by Mubayi and Suk, and by Conlon, Fox and Sudakov, for $k\ge4$ and $s\ge k+3$. In this paper, we prove that $r_4(6,n)\ge 2^{2^{cn}}$ for some absolute constant $c>0$, improving upon our previous bound. Consequently, we confirm the Erdős–Hajnal conjecture for $r_k(k+2,n)$ for all fixed $k\ge4$.

The vocabulary used to describe AI agents in governance contexts – learning, memory, values, compliance, identity, trust – is borrowed from psychological and organizational science, contributing to systematic failures in how organizations deploy, oversee, and hold agents accountable. This paper argues that the problem is not merely terminological but epistemological: psychological vocabulary carries an "invisible grammar" of its home discipline into governance discourse, calibrating frameworks to a metaphysical entity that does not exist in current AI architectures. We call this the disciplinary language transfer problem. Drawing on Wittgenstein's concept of language games, Kuhn's paradigm-laden observation, Haraway's situated knowledge, and Star and Griesemer's boundary object theory, we show that the transfer operates at three levels (epistemological assumptions, theoretical constructs, and surface vocabulary), each requiring a different remediation. We characterize six foundational epistemological assumptions embedded in Western psychological governance discourse, trace their origin in specific philosophical traditions, and show why each fails when applied to systems without developmental continuity. The paper's practical output is an actionable Disciplinary Audit: a six-question governance document scan operationalized through a translation taxonomy of thirty-seven terms mapping operational constructs to agent-appropriate replacements, presented here in abridged form and openly archived in full. The vocabulary reform proposed here is not merely terminological; it is the condition of possibility for governance frameworks that correctly identify what they are governing.

We present an improvement on previous spacecraft pose estimation architectures that results in the lowest published mean rotation errors we know of on the SPEED+ lightbox and sunlamp test sets for a known, non-cooperative spacecraft. By using a previously established heatmap-based pose estimation architecture and adapting a large self-supervised ViT foundation model (DINOv3) in place of the smaller convolutional and ViT encoders of previous work, we show that pose estimation accuracy improves from 300M to 840M parameters with no saturation yet observed. We also evaluate our 840M model on a Jetson Orin NX 16GB, measuring single-pass network inference at 133.8 ms per crop with a board draw of 32.0 W. These measurements demonstrate embedded inference feasibility on a processor family with orbital flight heritage. Our resulting model outperforms previous models across lightbox and sunlamp domains while training only on synthetic data. Our best model, using DINOv3 840M adapted with LoRA as the encoder (rank 64, three-seed ensemble with four-rotation test-time augmentation), results in $1.56^\circ$ mean rotation error on sunlamp and $1.17^\circ$ on lightbox, compared to the previous best mean rotation errors we know of on these test sets, $2.66^\circ$ and $1.75^\circ$ by EagerNet.

Magnitude is an isometric invariant of metric spaces introduced by Leinster in 2011. It is nowhere continuous on the Gromov–Hausdorff space of finite metric spaces, however, positive continuity results do exist if we restrict the ambient space. In this paper, we prove that magnitude is continuous at every finite subset $F$ of $\ell_1^N$. We do this by first deriving the weight measure for a finite union of cubes, i.e.\ cubical thickenings of the points in $F$, and then showing that these thickenings converge to the magnitude of the underlying finite set.

In this paper, we establish a general volume estimate for complete Riemannian manifolds under suitable differential inequalities involving a proper function and a symmetric tensor, without imposing curvature assumptions. As an application, we prove that a class of orientable hypersurfaces properly immersed in Euclidean space has at most polynomial volume growth and finite weighted volume. This class includes, in particular, certain self-shrinkers of the $Q_k$-flow and properly immersed minimal hypersurfaces for which the normal component of the position vector is bounded.

In 2015, Bondesson showed that Thorin's class of generalized gamma convolutions (GGCs) has the remarkable property of being closed under multiplication of independent random variables. He also conjectured that the GGC class is closed under taking powers of order greater than one. In this paper, we provide a candidate proof for Bondesson's conjecture.

In this paper, we argue that acceptance and rejection are classical speech acts that express two opposite polar positions toward propositional content and can be systematically represented within a classically outlined bilateral formal model. Acceptance occupies the positive pole, represented by positive A, whereas rejection occupies the negative pole, represented by minus A. Their classical relationship is mediated by negation because acceptance of A corresponds to rejection of negation A, while rejection of A corresponds to acceptance of negation A, represented by positive A iff minus negation A and minus A iff positive negation A. Coherency and incoherency are introduced as supplementary notions to clarify this relationship and its connection with reflexivity. A further result of the manuscript is the methodical distinction between rejection and negation namely, rejection expresses negative polarity toward A, whereas negation operates upon propositional content.

In wireless communication networks, many resource optimization problems are nondeterministic polynomial-time hard (NP-hard) due to their combinatorial nature and high computational complexity. Recently, neutral-atom-based quantum computing has emerged as a promising platform for efficiently solving such problems by leveraging quantum superposition and entanglement. However, its application to wireless communication optimization problems remains largely unexplored. In this paper, we investigate the use of neutral-atom quantum platforms to solve the maximum access problem (MAP), formulated as a mixed-integer programming task that jointly considers admission control, user clustering, channel assignment, and power allocation in a non-orthogonal multiple access (NOMA)-enabled uplink network. To reduce the computational burden, the MAP is equivalently reformulated as a maximum independent set (MIS) problem in graph theory. This reformulation enables the use of the neutral atom platform based on Rydberg atom arrays, where the MIS problem is naturally encoded into the physical geometry and blockade constraints of the quantum system. Numerical results demonstrate the feasibility and potential of this approach for addressing large-scale wireless resource optimization problems.

Honeypots are designed to deceive attackers, and recent work shows they can also derail autonomous LLM-based pentesters. These evaluations, however, largely consider attackers unaware of the deception they face. We study the opposite setting: an autonomous attacker explicitly equipped to recognize and act on honeypot fingerprints. We introduce Rouxii, an AI-driven penetration-testing framework that integrates counter-deception into reconnaissance and pivots from honeypot detection to exploitation. We evaluate matched vanilla and anti-deception Rouxii configurations across three reasoning models and eleven network setups over twelve cycles (1,544 attack reports). Between the matched cohorts, which differ only in the prompt, counter-deception raises correct honeypot identification from 19% to 97%, an effect strongest on OT services (11% to 97%), while false alarms on the real service stay at 0.7%. Deception-unaware baselines (PentestGPT, HackingBuddy) fail similarly, indicating the effect is not specific to our framework. Detection, moreover, is not the endpoint: through a white-box analysis of the honeypots themselves we show that a detected trap can be turned against its operator, demonstrating a denial-of-service that disables Conpot without tripping its liveness monitoring, and a corruption of the intelligence a GasPot instance reports. These findings show that deception effectiveness depends strongly on attacker knowledge, and that evaluations of honeypot resilience against AI attackers must account for adversaries that actively reason about and exploit the deception layer.

Superfluorescence (SF), intense picosecond bursts from self-synchronizing dipoles, is promising for room-temperature quantum sources. Yet, whether this synchronization can convert weak structural anisotropies into robust macroscopic order remains an open question. Here, we demonstrate that polycrystalline ferroelectric 2D perovskite thin films exhibit strongly linearly polarized SF with a degree of polarization (DOP) up to 86%. Remarkably, macroscopic polarization emerges without external bias despite randomly oriented microscopic domains, suggesting a cooperative gain mechanism that amplifies weak local anisotropies by several orders of magnitude into coherent, linearly polarized light bursts. Our results establish cooperative superfluorescence as a general concept for translating microscopic anisotropy into robust macroscopic order, exemplified here for linear polarization, offering a sensitive probe of hidden material symmetries and a framework for polarized quantum light sources

Primordial non-Gaussianity is a direct probe of the physics of inflation. The tightest bounds still come from the cosmic microwave background, but large-scale structure is rapidly catching up: its intrinsically three-dimensional maps contain far more modes, provided the tracer combines large volume, high redshift, strong bias and sufficient number density. The Little Red Dots (LRDs) discovered by JWST plausibly offer all four. Inspired by recent work on BAO measurements using LRDs we forecast the joint galaxy power spectrum and bispectrum of a $14{,}000\,{\rm deg}^2$ LRD spectroscopic survey over $4<z<9$, marginalising over $32$ nuisance parameters and including the non-Gaussian bispectrum covariance. We find $\sigma(f_{\mathrm{NL}}^{\mathrm{loc}})=0.32$, $\sigma(f_{\mathrm{NL}}^{\mathrm{equil}})=32$ and $\sigma(f_{\mathrm{NL}}^{\mathrm{orth}})=10$, improving on Planck by factors of $16$, $1.5$ and $2.3$. The local bound lies a factor of three below the $\sigma(f_{\mathrm{NL}}^{\mathrm{loc}})\simeq1$ threshold that separates broad classes of single- and multi-field inflation. LRD clustering therefore has the potential to become an important probe of inflation, and makes a strong case for future wide-field near- to mid-infrared spectroscopy.

Exponential Second-Order Expected Utility (SOEU) underlies the entropic approach to model uncertainty. This paper explores in what sense that functional form is essential. In the misspecification-robust Smooth Ambiguity criterion, let a single parameter govern both the model-level robustness and the ambiguity-averse aggregation across models Cerreia-Vioglio, Hansen, Maccheroni, and Marinacci (2026). The two-layer criterion then equals Exponential SOEU for every compact set of models and every second-order prior, with the Bayesian predictive measure as the baseline. The main results are converses. On the aggregator side, matched curvature is necessary: at a fixed curvature no other continuous, strictly increasing aggregator delivers the reduction, and with mismatched curvature there are a model set and a prior for which no single-layer entropic value, at any curvature and any baseline, reproduces the criterion. On the cost side, within the power-divergence family, which contains chi-squared and reverse Kullback–Leibler (KL), only KL has a dual of the log-sum-exp form, and the other members first depart from it at the third cumulant. Finally, the value dual to Exponential SOEU is a robust-control value: it is motivated by Rational Inattention, but no Bayes-plausible information-acquisition problem about a fixed act generates it.

Programmable RISC-V soft cores are becoming more widespread in data-center scenarios, making it crucial to design efficient systems that deploy them on FPGA chips with HBM memory. Toki, released as open source, is the first hardware-software framework that enables profiling the performance of HBM on FPGA accelerator cards by jointly considering (i) the execution of workloads on RISC-V soft cores instantiated on the FPGA and (ii) the injection of memory traffic from the host system via DMA over PCIe, providing insights that cannot be obtained with synthetic traffic generators alone. Extensive experiments target an AMD Alveo U55C card, deploying up to 60 RISC-V compute cores and stressing its HBM2 memory through real-world applications and microbenchmarks with user-defined access patterns. Results showcase how Toki can effectively profile the impact on HBM performance of the compute cores' organization, the workload's memory access patterns, data locality, contention over the memory controllers, and host traffic.

LLM-as-a-judge scales evaluation, but reasoning judges are slow and costly. We study JEV-as-a-Judge: evaluation with JEV, a decision-only judge that returns label probabilities instead of text, and whose confidence decides whether to accept its verdict or escalate to a reasoning judge. Against sixteen generative and reward-model judges, with blinded human adjudication, JEV comes within three points of GPT-6 wherever a verdict can be read off the text, at 0.36% of its fee and a 0.15-second median latency, and falls behind where the verdict must be derived, as in math, code, and logic. Its confidence marks this boundary. With a threshold frozen in advance, accepting confident verdicts and escalating the rest is 0.9 points more accurate than GPT-6 on 1,610 held-out pairs at 41% of its fee, and in a pre-specified live test on two new workloads the cascade matches GPT-6's accuracy exactly. Confidence routing weakens on style-adversarial pairs and reference-free prose; we close with a simple recipe for validating thresholds locally.

Individual tree crown segmentation from aerial imagery underpins tree-level carbon accounting, biodiversity, and restoration monitoring at landscape scale. However, existing models are predominantly trained on dense canopy forest imagery and degrade in savannah and drylands, where tree crowns are sparse, of variable appearance, and underrepresented in annotated benchmarks. These models also typically depend on costly polygon annotations. We introduce LACE (LAtent Commonality Expectation-maximisation), a box-supervised instance segmentation model, evaluated on 0.1 m/px aerial RGB tree crown imagery. LACE uses a frozen DINOv3-web ViT-L/16 encoder, applied at four spatial offsets and interlaced into a denser feature grid, with a lightweight CenterNet-style detection head trained solely on bounding boxes. We use expectation-maximisation to separate recurring appearance, the "treeness", within bounding boxes from surroundings. On the OAM-TCD benchmark test set, LACE reaches a mask AP$_{50}$ of $0.663 \pm 0.001$ (3 seeds) trained on 900 box-annotated images and without mask annotations, above the 0.626 scored by Restor's released mask-supervised Mask R-CNN, which was trained on the full ~4.2k image set. On a sparse-canopy holdout set, mask AP$_{50}$ rises to $0.691$ versus $0.612$ for Detectree2, a mask-supervised baseline. On NeonTreeEvaluation, using the official evaluation code, LACE reaches $0.728 \pm 0.003$ F1@0.4 (5 seeds) from 23,424 hand-annotated RGB boxes alone, matching the authors' DeepForest model's published 0.719, using under 0.1% of its training annotations and none of its LiDAR-derived 30M-crown pretraining set. By leveraging frozen self-supervised features, LACE matches or surpasses fully-supervised specialist baselines from boxes alone, removing the need for polygon annotation in tree crown instance segmentation for sparse-canopy environments where labelled data is scarce.

Probability backflow is investigated for charged spin-$1/2$ particles in a uniform magnetic field within the lowest-Landau-level approximation. While single-mode Landau states exhibit only weak and non-robust backflow, it is shown that two-mode interference gives rise to negative probability flux in both the Schrödinger and Pauli formulations. In the Pauli case, orbital and spin contributions to the probability current are explicitly separated, revealing an additional interference mechanism. The external magnetic field is found to act as a tunable control parameter, simultaneously controlling the spatial overlap of orbitals and the spin dynamics. In the two-mode regime, a resonance between orbital mismatch and spin precession leads to an enhancement of backflow. Finite spatial resolution is shown not to suppress the effect within realistic coarse-graining. These results establish magnetic-field-controlled quantum backflow in Landau systems as an accessible interference phenomenon.

Accurate generation of crystal structures is the foundation to the discovery of high-performance materials for extreme-environment applications, such as aerospace, additive manufacturing, and fusion energy systems. Although generative modeling has emerged as a promising approach for crystal design, its performance remains limited by the complex crystal structures and diverse chemical compositions. In this work, we develop UFO-MGen, a universal flow-based generative model that learns topological features of Wyckoff representations and leverages this information to accurately generate crystals across vast structural and chemical spaces. Compared with state-of-the-art generative models, UFO-MGen achieves the highest crystal generation success rate under a rigorous multi-stability evaluation framework, the highest SUN (stable, unique, novel) rate, and a remarkable extrapolation capability that has not been reported by previous models. Furthermore, a fine-tuning module is implemented to UFO-MGen for property-constrained crystal generation, enabling the inverse materials design toward target properties. The UFO-MGen opens a new avenue for accelerated materials discovery and providing a foundation for universal materials intelligence.

We exhibit simple, rotationally symmetric Venn diagrams with 17 curves, with 19 curves and with 23 curves: $n$ Jordan curves carried to one another by rotation through $2\pi/n$, with every one of the $2^n$ regions present and connected and, since the diagrams are simple, every crossing on exactly two curves. Symmetric Venn diagrams exist for every prime number of curves (Griggs, Killian and Savage, 2004), but those diagrams have many curves through a point; simple ones were known only up to 13 curves (Mamakani and Ruskey, 2014). Four 17-curve, twelve 19-curve and six 23-curve diagrams were found by a Metropolis walk on rotation-invariant quadrangulations of the sphere in which regions may temporarily be duplicated, started from the Griggs-Killian-Savage diagram with its multiple crossings resolved. At 23 curves the walk was held for two weeks by duplicated regions near the poles; the two lineages that finished were the first whose $E=92$ states had none. Every diagram is given by a machine-checkable certificate; one certificate each of the 17-, 19- and 23-curve sizes has been verified by a formal proof in Lean 4. All of the diagrams are non-monotone, which is why the crossing-sequence searches that found the 11- and 13-curve diagrams could not have found them.

We investigate the potential of Higgs cascade decays to probe neutralino dark matter below the neutrino floor in the semi-constrained Next-to-Minimal Supersymmetric Standard Model with nonuniversal gaugino masses. We focus on lightest-neutralino masses of 65 to 100 GeV and study the process $pp\to H_2\to A_1A_1\to\gamma\gamma+\tilde{\chi}_1^0\tilde{\chi}_1^0$, where $H_1$ is the observed SM-like Higgs boson, $H_2$ is a heavier CP-even state, and $A_1$ is the lightest CP-odd state. The selected parameter region lies near the $A_1$ resonance, with $m_{A_1}>2m_{\tilde{\chi}_1^0}$, allowing both resonantly enhanced neutralino annihilation and invisible $A_1$ decays. After imposing theoretical and experimental constraints, we find surviving samples with cascade signal cross sections of approximately 0.16 fb at $\sqrt{s}=14~\mathrm{TeV}$ even when their relic-density-rescaled spin-independent scattering cross sections lie below the neutrino floor. A Monte Carlo analysis of a representative benchmark point yields an expected statistical significance of approximately $6.1\sigma$ at the high-luminosity LHC with an integrated luminosity of $3000~\mathrm{fb}^{-1}$, neglecting systematic uncertainties. These results highlight the potential of the diphoton plus missing transverse momentum channel as a complementary probe of neutralino dark matter.

Simulating strongly correlated fermions remains a long-standing challenge due to the exponential complexity of the Hilbert space and the intricate sign structure of many-body wavefunctions. We introduce a variational framework centered on a neural network backflow transformation that combines deep learning with variational Monte Carlo. The proposed ansatz employs a multi-determinant expansion with low-rank shifts to capture non-local correlations and complex sign structures. Applied to the two-dimensional Hubbard model at both half-filling and $1/8$ doping, the method achieves energies within $0.45\%$ of auxiliary-field quantum Monte Carlo at half-filling and captures intertwined charge- and spin-density stripe patterns at $1/8$ doping. These results demonstrate the potential of this framework as a scalable and interpretable approach to variational simulations of strongly correlated fermionic systems.

We study a social contagion model describing the propagation of opinions across a network. Our model is a generalization of a model introduced by Daley and Kendall, and consists of a system of coupled differential equations set on the nodes of an infinite graph. The modeling is inspired by SIR-type models from epidemiology. The key idea is that individuals can transmit their opinion to their neighbors (as in epidemiological models), until they perceive that the opinion is already known in their neighborhood. After introducing the model, we analyze the long-time behavior of its solutions. We prove that the solutions converge and we study their asymptotic equilibrium. We also establish estimates on the speed of propagation of the solutions through the network. These results give estimates on the rate of adoption of the opinion and on its velocity in the social network. We also provide simulations that indicate that this model can give rise to patterns: when the individuals take into account individuals that are ”far enough” in estimating whether or not the opinion is already known, then we observe the emergence of localized ”bubbles” where the opinion is adopted.

An Elastic Degenerate String (EDS, or ED-string) is a sequence of string sets. A pangenome, consisting of variations observed in a population along the genome sequences, can be naturally encoded as an EDS. Pattern matching and comparison problems on pangenome representations such as EDSes have been widely studied in the literature, but optimizing the pangenome properties during its construction has been largely omitted. We fill this gap by showing how methods originally developed for the related problem of founder reconstruction can be adapted to minimize, in linear time, the total cardinality of the EDS sets or the total size of the EDS strings, given suitable multiple alignments representing the input data. We provide an implementation for the minimum-cardinality criterion in a tool mincard, and conduct the first experiments on scalable pangenome optimization via EDSes. The code and experiments are available at https://github.com/algbio/eds.

Every Hurwitz-stable interval family of monic real polynomials of degree six admits a single real numerator of degree six that makes all the associated transfer functions strictly positive real. We give a constructive proof. The complete existence theorem has been formalized in Lean4.

Lifelong learners from different industries often watch the same recorded lecture, even when they will apply the material in different workplaces. We present Bespoke, a system that takes the transcript of an existing lecture and generates a new video customized to a target industry and duration, with new slides, narration, and charts. From 31 graduate lectures in analytics, machine learning, and optimization, we generated 209 videos for these three industries, plus a generic-audience version of each lecture. Twenty-five experts in the corresponding domains rated 92 videos on a five-point rubric. They judged 87% at or above the rubric midpoint corresponding to “a standard MOOC lecture's quality” (mean overall score 3.42 out of 5; 48% scored 4 or above). Overall quality was similar across industries, durations, and a 21-lecture held-out set unused while developing the system, at about \$0.22 API cost per minute of video.

Large language models (LLMs) have increasingly been used to investigate how children acquire syntax at an early stage of development. This is notably the central scientific goal of the BabyLM challenge, a community-wide effort to develop models that achieve human-level syntactic performance while being trained on developmentally realistic corpora. In this paper, we reflect on the use of LLMs in the study of infant syntax learning by providing an epistemological assessment of several studies from this research program. We discuss how datasets are built, which models are implemented, how they are trained and syntactically evaluated. We observe significant assumptions in the methodology of BabyLM and related studies, thus mitigating their theoretical scope. We additionally observe that using developmentally-realistic corpora have limited effects on models performance on commonly-used benchmarks, which suggest important computational differences between LLMs and the infant syntax learner.

Cosmogenic $^{9}$Li and $^{8}$He isotopes constitute an important correlated background in low-energy neutrino experiments because their $\beta$-delayed neutron decay signatures can mimic inverse beta decay signals. Conventional estimates based on the time since the last muon become challenging at high muon rates, while muon-related vetoes further reduce the residual isotope statistics. We extend the conventional time fit to an event-level muon-categorized joint time (J-MuCAT) fit, which uses the time to the most recent preceding muon in each energy-loss category. A complementary estimate is obtained from the candidate-to-muon-track distance distribution (TraDiTS). The muon-uncorrelated component is determined from the far-distance region and subtracted. This estimate is then used to constrain the J-MuCAT fit, defining the distance-constrained J-MuCAT (DCJ-MuCAT) fit. In detector-level simulation with successive cosmogenic-background vetoes, DCJ-MuCAT reduces the statistical uncertainty by more than $40\%$ relative to J-MuCAT and by more than $10\%$ relative to the TraDiTS. The fitted results remain consistent with the simulation truth. Applicability studies further show good performance over a broad range of muon rates and isotope fractions. The proposed framework provides a practical approach for estimating residual $^{9}$Li/$^{8}$He backgrounds in large neutrino detectors.

These notes develop the mathematical foundations and construction of a Fourier-Bessel wavelet family inspired by the disk harmonics of Shaqfa et al.[9]. We begin with the relevant properties of Bessel and modified Bessel functions and introduce the wavelet properties required for the construction. We then derive the Fourier-Bessel disk harmonics as solutions to the Helmholtz equation on the unit disk subject to a Neumann boundary condition. Building on this basis, we construct a wavelet family by applying a Gaussian spatial envelope and introducing a zero-mean correction for the zeroth angular order. We derive the corresponding normalisation constants for $L^2$-based applications and discuss $L^1$-based normalisation for frequency-domain peak consistency. Finally, we derive a closed-form Fourier-domain representation of the resulting wavelets. The main motivation is the approximately linear spacing, which converges to $\pi$ between consecutive radial eigenvalues. Rather than replacing the conventional dyadic organisation of wavelet families, this construction lays out the foundation to explore whether a more uniform radial frequency allocation can be useful for applications in which broad and balanced frequency coverage is desirable.

In LLM-based speech translation, transcription-based chain-of-thought (CoT) suffers from a mismatch between reference transcripts used in supervised fine-tuning (SFT) and model-generated transcripts at inference. To address this, we propose joint recognition and translation fine-tuning via group relative policy optimization (GRPO). We score both transcripts and translations, with translation conditioned on model-generated transcripts, and compare three token advantage strategies. Using Qwen2.5-Omni-3B across four languages, we evaluate CoT against direct speech translation (Direct ST) under SFT and GRPO, training on CoVoST 2 and testing on CoVoST 2 and FLEURS. CoT GRPO outperforms Direct ST GRPO by 1.77 and 0.83 average BLEU points on CoVoST 2 and FLEURS. Compared to CoT SFT, GRPO boosts BLEU by 0.82 and 0.67 points and reduces word error rate (WER) by 8.8% and 7.2% relatively. These results highlight reinforcement fine-tuning as an effective method to mitigate the training-inference mismatch, jointly improving recognition and translation.

A catalogue (https://doi.org/10.5281/zenodo.20157853) of solar radio bursts detected with the LV614 LOFAR station at the Irbene Observatory, Latvia, in the stand-alone mode from June 2022 to August 2025 in the 29-60 MHz frequency range is presented. Observations were conducted with the weekly cadence, one-second time resolution, and 0.195 MHz spectral resolution. The cumulative duration of the observation was 1125.2 h. For all analysed observations, bandpass calibration was applied using Cassiopeia A as a reference source. In total, the catalogue includes 335 solar radio bursts. Radio bursts of all five main types, I-V, were detected. Specifically, the catalogue contains parameters of 23, 8, 293, 6, and 5 bursts of Type I, II, III, IV, and V, respectively. The detected burst peak flux ranges from 0.3 SFU to 234 SFU. The catalogue provides information about the peak flux and frequency, and the estimated height of the source above the solar surface. For Type II and III bursts, we also give the estimations of the frequency drifts, with mean values $ 0.043 ~\mathrm{MHz\,s^{-1}}$ and $4.383 ~\mathrm{MHz\,s^{-1}}$, respectively. We also estimated the fit $df/dt = Af^a$, with the best fit parameters $A=0.78$ and $a=0.43$. The numbers of Type I and III events are found to decrease with the peak flux, with the power law indices about $-0.9$ and $-1.25$, respectively. For Type III, we estimated electron beam speed ranging from 0.005 to 0.477 of the speed of light for the fundamental frequency, and 0.008 to 0.794 of the speed of light for its second harmonic.