Search

Domain wall (DW) dynamics in antiferromagnetic (AFM) systems offer the advantages over their ferromagnetic counterparts of having faster and more energy efficient manipulation due to the absence of net magnetization, leading to reduced magnetic crosstalk and improved performance in spintronic devices. A comprehensive analysis of DW dynamics across regimes such as creep, depinning, and flow is well established in ferromagnetic systems but remains lacking in AFM-coupled systems. In this study, we explore the nature of DW dynamics in synthetic ferrimagnetic multilayers composed of Pt|Co|Tb|Al for different Tb thickness, focusing on the underlying pinning parameters, and on the different regimes of DW dynamics driven by spin-orbit torques (SOTs). We find that due to the AFM coupling between Co and Tb, the magnetic moment of Tb increases with Tb thickness resulting in a reduced saturation magnetization and an enhanced depinning field. The DW disorder interaction is found to vary weakly with the AFM coupling between Co and Tb, while the complete withdrawal of the Tb layer strongly increases the anisotropy and the DW pinning. Furthermore, we propose a novel approach to measure effective SOTs by comparing depinning transitions in current and field-induced DW motion. This research reveals new insights into DW dynamics in coupled AFM systems, highlighting enhancements in mobility through optimized SOTs and pinning landscapes., Comment: 12 pages, 9 figures

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,700 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai., Comment: 27 pages, 6 figures

Many active galaxies harbor powerful relativistic jets, however, the detailed mechanisms of their formation and acceleration remain poorly understood. To investigate the area of jet acceleration and collimation with the highest available angular resolution, we study the innermost region of the bipolar jet in the nearby low-ionization nuclear emission-line region (LINER) galaxy NGC 1052. We combined observations of NGC 1052 taken with VLBA, GMVA, and EHT over one week in the spring of 2017. For the first time, NGC 1052 was detected with the EHT, providing a size of the central region in-between both jet bases of 250 RS (Schwarzschild radii) perpendicular to the jet axes. This size estimate supports previous studies of the jets expansion profile which suggest two breaks of the profile at around 300 RS and 10000 RS distances to the core. Furthermore, we estimated the magnetic field to be 1.25 Gauss at a distance of 22 {\mu}as from the central engine by fitting a synchrotron-self absorption spectrum to the innermost emission feature, which shows a spectral turn-over at about 130 GHz. Assuming a purely poloidal magnetic field, this implies an upper limit on the magnetic field strength at the event horizon of 26000 Gauss, which is consistent with previous measurements. The complex, low-brightness, double-sided jet structure in NGC 1052 makes it a challenge to detect the source at millimeter (mm) wavelengths. However, our first EHT observations have demonstrated that detection is possible up to at least 230 GHz. This study offers a glimpse through the dense surrounding torus and into the innermost central region, where the jets are formed. This has enabled us to finally resolve this region and provide improved constraints on its expansion and magnetic field strength., Comment: 22 pages, 10 figures, published in A&A

The 2017 observing campaign of the Event Horizon Telescope (EHT) delivered the first very long baseline interferometry (VLBI) images at the observing frequency of 230 GHz, leading to a number of unique studies on black holes and relativistic jets from active galactic nuclei (AGN). In total, eighteen sources were observed: the main science targets, Sgr A* and M87 along with various calibrators. We investigated the morphology of the sixteen AGN in the EHT 2017 data set, focusing on the properties of the VLBI cores: size, flux density, and brightness temperature. We studied their dependence on the observing frequency in order to compare it with the Blandford-K\"onigl (BK) jet model. We modeled the source structure of seven AGN in the EHT 2017 data set using linearly polarized circular Gaussian components and collected results for the other nine AGN from dedicated EHT publications, complemented by lower frequency data in the 2-86 GHz range. Then, we studied the dependences of the VLBI core flux density, size, and brightness temperature on the frequency measured in the AGN host frame. We compared the observations with the BK jet model and estimated the magnetic field strength dependence on the distance from the central black hole. Our results indicate a deviation from the standard BK model, particularly in the decrease of the brightness temperature with the observing frequency. Either bulk acceleration of the jet material, energy transfer from the magnetic field to the particles, or both are required to explain the observations.

One-class anomaly detection aims to detect objects that do not belong to a predefined normal class. In practice training data lack those anomalous samples; hence state-of-the-art methods are trained to discriminate between normal and synthetically-generated pseudo-anomalous data. Most methods use data augmentation techniques on normal images to simulate anomalies. However the best-performing ones implicitly leverage a geometric bias present in the benchmarking datasets. This limits their usability in more general conditions. Others are relying on basic noising schemes that may be suboptimal in capturing the underlying structure of normal data. In addition most still favour the image domain to generate pseudo-anomalies training models end-to-end from only the normal class and overlooking richer representations of the information. To overcome these limitations we consider frozen yet rich feature spaces given by pretrained models and create pseudo-anomalous features with a novel adaptive linear feature perturbation technique. It adapts the noise distribution to each sample applies decaying linear perturbations to feature vectors and further guides the classification process using a contrastive learning objective. Experimental evaluation conducted on both standard and geometric bias-free datasets demonstrates the superiority of our approach with respect to comparable baselines. The codebase is accessible via our public repository., Comment: Published in WACV 2025

In the framework of the MACIV project, a consortium of French laboratories has deployed a temporary seismic network of 100 broadband stations in the French Massif Central (FMC) for 3-4 years (2023-2027). The project aims at imaging the crust and upper mantle of the FMC to better assess the sources of volcanism, and the impacts of the Variscan inheritance or the Cenozoic rift system on volcanic systems. A large-scale array of 35 broadband stations covers the entire FMC and complements the permanent networks to reach a homogeneous coverage with ~35 km spacing. This network, with XP code, is the French contribution to AdriaArray. The XP array is complemented with 3 quasi-linear north-south, east-west and northwest-southeast profiles with inter-station spacing of 5-20 km, making up the XF network of 65 stations. The profiles cross volcanic areas and the main Variscan structures. We describe the experimental setup designed to optimize the performance/cost ratio and minimize the number of field visits, the deployment, the state-of-health monitoring, the data management and the data quality control strategies, outcomes of our 15-years' experience with major temporary seismic experiments in France and neighboring countries, including AlpArray. We also show some preliminary results including hypocenter locations and receiver function analysis. The 2 broadband arrays will be supplemented in 2025 by a month-long deployment of 3 large-N dense arrays of 625 3-C short-period nodes. These dense arrays will complete our multi-scale seismic experiment and illuminate active faults and possible plumbing systems of the youngest volcanoes.

We study streaming algorithms for proportionally fair clustering, a notion originally suggested by Chierichetti et. al. (2017), in the sliding window model. We show that although there exist efficient streaming algorithms in the insertion-only model, surprisingly no algorithm can achieve finite multiplicative ratio without violating the fairness constraint in the sliding window. Hence, the problem of fair clustering is a rare separation between the insertion-only streaming model and the sliding window model. On the other hand, we show that if the fairness constraint is relaxed by a multiplicative $(1+\varepsilon)$ factor, there exists a $(1 + \varepsilon)$-approximate sliding window algorithm that uses $\text{poly}(k\varepsilon^{-1}\log n)$ space. This achieves essentially the best parameters (up to degree in the polynomial) provided the aforementioned lower bound. We also implement a number of empirical evaluations on real datasets to complement our theoretical results., Comment: ICLR 2025

We give an active learning algorithm for deterministic one-counter automata (DOCAs) where the learner can ask the teacher membership and minimal equivalence queries. The algorithm called OL* learns a DOCA in time polynomial in the size of the smallest DOCA, recognising the target language. All existing algorithms for learning DOCAs, even for the subclasses of deterministic real-time one-counter automata (DROCAs) and visibly one-counter automata (VOCAs), in the worst case, run in exponential time with respect to the size of the DOCA under learning. Furthermore, previous learning algorithms are ``grey-box'' algorithms relying on an additional query type - counter value query - where the teacher returns the counter value reached on reading a given word. In contrast, our algorithm is a ``black-box'' algorithm. It is known that the minimisation of VOCAs is NP-hard. However, OL* can be used for approximate minimisation of DOCAs. In this case, the output size is at most polynomial in the size of a minimal DOCA., Comment: 29 pages, 9 figures, 4 Algorithms

Software developing small and medium enterprises (SMEs) play a crucial role as suppliers to larger corporations and public administration. It is therefore necessary for them to be able to demonstrate that their products meet certain security criteria, both to gain trust of their customers and to comply to standards that demand such a demonstration. In this study we have investigated ways for SMEs to demonstrate their security when operating in a business-to-business model, conducting semi-structured interviews (N=16) with practitioners from different SMEs in Denmark and validating our findings in a follow-up workshop (N=6). Our findings indicate five distinctive security demonstration approaches, namely: Certifications, Reports, Questionnaires, Interactive Sessions and Social Proof. We discuss the challenges, benefits, and recommendations related to these approaches, concluding that none of them is a one-size-fits all solution and that more research into relative advantages of these approaches and their combinations is needed., Comment: This paper has been accepted to CHI 25

Protecting online privacy requires users to engage with and comprehend website privacy policies, but many policies are difficult and tedious to read. We present the first qualitative user study on Large Language Model (LLM)-driven privacy policy assessment. To this end, we build and evaluate an LLM-based privacy policy assessment browser extension, which helps users understand the essence of a lengthy, complex privacy policy while browsing. The tool integrates a dashboard and an LLM chat. In our qualitative user study (N=22), we evaluate usability, understandability of the information our tool provides, and its impacts on awareness. While providing a comprehensible quick overview and a chat for in-depth discussion improves privacy awareness, users note issues with building trust in the tool. From our insights, we derive important design implications to guide future policy analysis tools., Comment: 6 pages without appendices and references, 12 pages total, 3 figures, poster at CHI 2025. arXiv admin note: text overlap with arXiv:2501.16033

In this paper we address the issue of grounding for experiential concepts. Given that perceptual demonstratives are a basic form of such concepts, we examine ways of fixing the referents of such demonstratives. To avoid 'encodingism', that is, relating representations to representations, we postulate that the process of reference fixing must be bottom-up and nonconceptual, so that it can break the circle of conceptual content and touch the world. For that purpose, an appropriate causal relation between representations and the world is needed. We claim that this relation is provided by spatial and object-centered attention that leads to the formation of object files through the function of deictic acts. This entire causal process takes place at a pre-conceptual level, meeting the requirement for a solution to the grounding problem. Finally we claim that our account captures fundamental insights in Putnam's and Kripke's work on "new" reference.

Chevyrev-Galbraith and Goren-Love show that the successive minima of the Gross lattice of a supersingular elliptic curve can be used to characterize the endomorphism ring of that curve. In this paper, we show that the third successive minimum $D_3$ of the Gross lattice gives necessary and sufficient conditions for the curve to be defined over the field $\mathbb{F}_p$ or over the field $\mathbb{F}_{p^2}$. In the case where the curve $E$ is defined over $\mathbb{F}_p$, the value of $D_3$ can even yield finer information about the endomorphism ring of $E$., Comment: 24 pages, code available at https://github.com/gkorpal/minimal-gross

We consider branching processes for structured populations: each individual is characterized by a type or trait which belongs to a general measurable state space. We focus on the supercritical recurrent case, where the population may survive and grow and the trait distribution converges. The branching process is then expected to be driven by the positive triplet of first eigenvalue problem of the first moment semigroup. Under the assumption of convergence of the renormalized semigroup in weighted total variation norm, we prove strong convergence of the normalized empirical measure and non-degeneracy of the limiting martingale. Convergence is obtained under an Llog L condition which provides a Kesten-Stigum result in infinite dimension and relaxes the uniform convergence assumption of the renormalized first moment semigroup required in the work of Asmussen and Hering in 1976. The techniques of proofs combine families of martingales and contraction of semigroups and the truncation procedure of Asmussen and Hering. We also obtain L^1 convergence of the renormalized empirical measure and contribute to unifying different results in the literature. These results greatly extend the class of examples where a law of large numbers applies, as we illustrate it with absorbed branching diffusion, the house of cards model and some growth-fragmentation processes.

4D flow MRI allows for the estimation of three-dimensional relative pressure fields, providing rich pressure information, unlike catheterization and Doppler echocardiography, which provide one-dimensional pressure drops only. The accuracy of one-dimensional pressure drops derived from 4D flow has been explored in previous literature, but additional work must be done to evaluate the accuracy of three-dimensional relative pressure fields. This work presents an analysis of three state-of-the-art relative pressure estimators: virtual Work-Energy Relative Pressure (vWERP), the Pressure Poisson Estimator (PPE), and the Stokes Estimator (STE). Spatiotemporal behavior and sensitivity to noise were determined in silico. Estimators were validated with a type B aortic dissection (TBAD) flow phantom with varying tear geometry and an array of twelve catheter pressure measurements. Finally, the performance of each estimator was evaluated across eight patient cases. In silico pressure field errors were lower in STE compared to PPE, although PPE pressures were less affected by noise. High velocity gradients and low spatial resolution contributed most significantly to local variations in 3D error fields. Low temporal resolution leads to highly transient peak pressure events being averaged, systematically underestimating peak pressures. In the flow phantom analysis, vWERP was the most accurate method, followed by STE and PPE. Each pressure estimator strongly correlated with ground truth pressure values despite the tendency to underestimate peak pressures. Patient case results demonstrated that the pressure estimators could be feasibly integrated into a clinical workflow., Comment: 10 pages, 8 figures. Planned submission to IEEE Transactions on Medical Imaging

The branching fraction of the decay $B^+\to \psi(2S)\phi(1020)K^+$, relative to the topologically similar decay $B^+\to J/\psi \phi(1020) K^+$, is measured using proton-proton collision data collected by the LHCb experiment at center-of-mass energies of 7, 8, and 13 TeV, corresponding to an integrated luminosity of $9\,\mathrm{fb}^{-1}$. The ratio is found to be $0.061 \pm 0.004 \pm 0.009$, where the first uncertainty is statistical and the second systematic. Using the world-average branching fraction for $B^+ \to J/\psi \phi(1020) K^+$, the branching fraction for the decay $B^+\to \psi(2S) \phi(1020) K^+$ is found to be $ (3.0 \pm 0.2 \pm 0.5 \pm 0.2) \times 10^{-6}$, where the first uncertainty is statistical, the second systematic, and the third is due to the branching fraction of the normalization channel., Comment: All figures and tables, along with machine-readable versions and any supplementary material and additional information, are available at https://lbfence.cern.ch/alcm/public/analysis/full-details/3320/ (LHCb public pages)

SPARQL query editors often lack intuitive interfaces to aid SPARQL-savvy users to write queries. To address this issue, we propose an easy-to-deploy, triple store-agnostic and open-source query editor that offers three main features: (i) automatic query example rendering, (ii) precise autocomplete based on existing triple patterns including within SERVICE clauses, and (iii) a data-aware schema visualization. It can be easily set up with a custom HTML element. The tool has been successfully tested on various public endpoints, and is deployed online at https://sib-swiss.github.io/sparql-editor with open-source code available at https://github.com/sib-swiss/sparql-editor.

Pinching-antenna systems (PASSs) are a recent flexible-antenna technology that is realized by attaching simple components, referred to as pinching elements, to dielectric waveguides. This work explores the potential of deploying PASS for uplink and downlink transmission in multiuser MIMO settings. For downlink PASS-aided communication, we formulate the optimal hybrid beamforming, in which the digital precoding matrix at the access point and the location of pinching elements on the waveguides are jointly optimized to maximize the achievable weighted sum-rate. Invoking fractional programming and Gauss-Seidel approach, we propose two low-complexity algorithms to iteratively update the precoding matrix and activated locations of the pinching elements. We further study uplink transmission aided by a PASS, where an iterative scheme is designed to address the underlying hybrid multiuser detection problem. We validate the proposed schemes through extensive numerical experiments. The results demonstrate that using a PASS, the throughput in both uplink and downlink is boosted significantly as compared with baseline MIMO architectures, such as massive MIMO~and classical hybrid analog-digital designs. This highlights the great potential of PASSs, making it a promising reconfigurable antenna technology for next-generation wireless systems., Comment: 13 pages and 10 figures

We show that the theories of some (ordered) central simple algebras with involution over real closed fields are model-complete or admit quantifier elimination, and characterize positive cones in terms of morphisms into models of some of these theories.

We have explored the temporal variability of the seismicity at global scale over the last 124 years, as well as its potential drivers. To achieve this, we constructed and analyzed an averaged global seismicity curve for earthquakes of magnitude equal or greater than 6 since 1900. Using Singular Spectrum Analysis, we decomposed this curve and compared the extracted pseudo-cycles with two global geophysical parameters associated with Earth's tides: length-of-day variations and sea-level changes. Our results reveal that these three geophysical phenomena can be be explained with 90% accuracy, as the sum of up to seven periodic components, largely aligned with planetary ephemerides: 1 year, 3.4 years (Quasi-Biennial Oscillation, QBO), $\sim$11 years, $\sim$14 years, $\sim$18.6 years (lunar nodal cycle), $\sim$33 years, and $\sim$60 years. We discuss these results in the framework of Laplace's theory, with a particular focus on the phase relationships between seismicity, length-of-day variations, and sea-level changes to further elucidate the underlying physical mechanisms. Finally,integrating observations from seismogenic regions, we propose a trigger mechanism based on solid Earth-hydrosphere interactions, emphasizing the key role of water-rock interactions in modulating earthquake occurrence., Comment: 37 pages, 11 Figures

As Large Langue Models have been shown to memorize real-world facts, the need to update this knowledge in a controlled and efficient manner arises. Designed with these constraints in mind, Knowledge Editing (KE) approaches propose to alter specific facts in pretrained models. However, they have been shown to suffer from several limitations, including their lack of contextual robustness and their failure to generalize to logical implications related to the fact. To overcome these issues, we propose SAKE, a steering activation method that models a fact to be edited as a distribution rather than a single prompt. Leveraging Optimal Transport, SAKE alters the LLM behavior over a whole fact-related distribution, defined as paraphrases and logical implications. Several numerical experiments demonstrate the effectiveness of this method: SAKE is thus able to perform more robust edits than its existing counterparts.

DUSt3R introduced a novel paradigm in geometric computer vision by proposing a model that can provide dense and unconstrained Stereo 3D Reconstruction of arbitrary image collections with no prior information about camera calibration nor viewpoint poses. Under the hood, however, DUSt3R processes image pairs, regressing local 3D reconstructions that need to be aligned in a global coordinate system. The number of pairs, growing quadratically, is an inherent limitation that becomes especially concerning for robust and fast optimization in the case of large image collections. In this paper, we propose an extension of DUSt3R from pairs to multiple views, that addresses all aforementioned concerns. Indeed, we propose a Multi-view Network for Stereo 3D Reconstruction, or MUSt3R, that modifies the DUSt3R architecture by making it symmetric and extending it to directly predict 3D structure for all views in a common coordinate frame. Second, we entail the model with a multi-layer memory mechanism which allows to reduce the computational complexity and to scale the reconstruction to large collections, inferring thousands of 3D pointmaps at high frame-rates with limited added complexity. The framework is designed to perform 3D reconstruction both offline and online, and hence can be seamlessly applied to SfM and visual SLAM scenarios showing state-of-the-art performance on various 3D downstream tasks, including uncalibrated Visual Odometry, relative camera pose, scale and focal estimation, 3D reconstruction and multi-view depth estimation., Comment: Accepted at CVPR 2025

Recent works have successfully demonstrated that sparse deep reinforcement learning agents can be competitive against their dense counterparts. This opens up opportunities for reinforcement learning applications in fields where inference time and memory requirements are cost-sensitive or limited by hardware. Until now, dense-to-sparse methods have relied on hand-designed sparsity schedules that are not synchronized with the agent's learning pace. Crucially, the final sparsity level is chosen as a hyperparameter, which requires careful tuning as setting it too high might lead to poor performances. In this work, we address these shortcomings by crafting a dense-to-sparse algorithm that we name Eau De $Q$-Network (EauDeQN). To increase sparsity at the agent's learning pace, we consider multiple online networks with different sparsity levels, where each online network is trained from a shared target network. At each target update, the online network with the smallest loss is chosen as the next target network, while the other networks are replaced by a pruned version of the chosen network. We evaluate the proposed approach on the Atari $2600$ benchmark and the MuJoCo physics simulator, showing that EauDeQN reaches high sparsity levels while keeping performances high., Comment: Accepted at RLDM

A growing body of work in Ethical AI attempts to capture human moral judgments through simple computational models. The key question we address in this work is whether such simple AI models capture {the critical} nuances of moral decision-making by focusing on the use case of kidney allocation. We conducted twenty interviews where participants explained their rationale for their judgments about who should receive a kidney. We observe participants: (a) value patients' morally-relevant attributes to different degrees; (b) use diverse decision-making processes, citing heuristics to reduce decision complexity; (c) can change their opinions; (d) sometimes lack confidence in their decisions (e.g., due to incomplete information); and (e) express enthusiasm and concern regarding AI assisting humans in kidney allocation decisions. Based on these findings, we discuss challenges of computationally modeling moral judgments {as a stand-in for human input}, highlight drawbacks of current approaches, and suggest future directions to address these issues., Comment: In ACM Conference on Human Factors in Computing Systems (CHI), 2025

The rise in complexity of network data in neuroscience, social networks, and protein-protein interaction networks has been accompanied by several efforts to model and understand these data at different scales. A key multiscale network modeling technique posits hierarchical structure in the network, and by treating networks as multiple levels of subdivisions with shared statistical properties we can efficiently discover smaller subgraph primitives with manageable complexity. One such example of hierarchical modeling is the Hierarchical Stochastic Block Model, which seeks to model complex networks as being composed of community structures repeated across the network. Incorporating repeated structure allows for parameter tying across communities in the SBM, reducing the model complexity compared to the traditional blockmodel. In this work, we formulate a framework for testing for repeated motif hierarchical structure in the stochastic blockmodel framework. We describe a model which naturally expresses networks as a hierarchy of sub-networks with a set of motifs repeating across it, and we demonstrate the practical utility of the test through theoretical analysis and extensive simulation and real data.

It is known that the unique ergodicity of the viscous primitive equations with additive white-in-time noise remains an open problem. In this work, we demonstrate that, as the rotational intensity approaches infinity, the distribution of any given strong solution, rescaled by the rotation, is attracted to the unique invariant measure of the stochastic limit resonant system. This suggests that the system exhibits nearly unique ergodicity in the large rotation limit. The proof is based on a stochastic averaging principle combined with a coupling argument., Comment: 44 pages

Classical Cepheids are not only excellent standard candles, but also invaluable tools to test stellar evolution and pulsation theories. Rates of their pulsation period change, quantified usually through $O-C$ diagrams, can be confronted with predictions of stellar evolution theory. On the other hand, period changes on much shorter time scales ($\sim$10$^{2}$-10$^{4}$days), attributed to non-evolutionary effects are often detected and lack detailed explanation. We aim to provide a systematic and quantitative description of irregular or non-linear period changes in Cepheids. We aim to provide a systematic and quantitative description of irregular or non-linear period changes in Cepheids. We analysed part of the OGLE data for classical Cepheids in the Magellanic Clouds (MCs; from both Large Magellanic Cloud, LMC, and the Small Magellanic Cloud, SMC) using the modified Hertzsprung $O-C$ technique. A sample of 3658 stars, with the best quality data and void of additional low-amplitude periodicities (e.g. due to non-radial pulsations), that could impact the results, was selected for analysis. Based on $O-C$ shapes, stars were classified into three categories: no period change (class 1), linear period change (class 2), and irregular change (class 3). In our investigation, $33.5\pm0.7$\% of analysed stars show irregular period changes. Considering the pulsation mode, irregular period changes were detected in $16.5\pm0.7$\% of the analysed fundamental mode stars and in $68.1\pm1.2$\% of the first overtone stars. The amplitude of variability in the $O-C$ diagrams increases with the pulsation period, and at a given pulsation period, it is larger for first overtone stars. While the increase is linear for first overtone stars, for fundamental mode stars it becomes steeper as the pulsation period increases., Comment: 20 pages, 23 figures, 4 tables (including Appendix); Accepted for publication in Astronomy and Astrophysics

A key challenge in AI alignment is guiding large language models (LLMs) to follow desired behaviors at test time. Activation steering, which modifies internal model activations during inference, offers a potential solution. However, prior work in dense activation spaces struggles with superposition, wherein multiple features become entangled, limiting interpretability and precise control. In contrast, sparse representations provide an untapped opportunity for more interpretable behavior modulation. In this work, we introduce sparse activation steering (SAS), a method that leverages sparse autoencoders (SAEs) to steer LLM behavior in sparse spaces. By isolating behavior-specific features through a contrastive prompt-pairing approach, we define a set of features that can selectively reinforce or suppress behaviors. Experiments on Gemma 2 LLMs show that SAS vectors enable nuanced behavioral modulation and finer-grained control. Furthermore, scaling SAEs improves monosemanticity of SAS vectors, suggesting more reliable and interpretable interventions.

Traditional approaches to learning fair machine learning models often require rebuilding models from scratch, generally without accounting for potentially existing previous models. In a context where models need to be retrained frequently, this can lead to inconsistent model updates, as well as redundant and costly validation testing. To address this limitation, we introduce the notion of controlled model debiasing, a novel supervised learning task relying on two desiderata: that the differences between new fair model and the existing one should be (i) interpretable and (ii) minimal. After providing theoretical guarantees to this new problem, we introduce a novel algorithm for algorithmic fairness, COMMOD, that is both model-agnostic and does not require the sensitive attribute at test time. In addition, our algorithm is explicitly designed to enforce minimal and interpretable changes between biased and debiased predictions -a property that, while highly desirable in high-stakes applications, is rarely prioritized as an explicit objective in fairness literature. Our approach combines a concept-based architecture and adversarial learning and we demonstrate through empirical results that it achieves comparable performance to state-of-the-art debiasing methods while performing minimal and interpretable prediction changes.

Gene drive alleles bias their own inheritance to offspring. They can fix in a wild-type population in spite of a fitness cost, and even lead to the eradication of the target population if the fitness cost is high. However, this outcome may be prevented or delayed if areas previously cleared by the drive are recolonised by wild-type individuals. Here, we investigate the conditions under which these stochastic wild-type recolonisation events are likely and when they are unlikely to occur in one spatial dimension. More precisely, we examine the conditions ensuring that the last individual carrying a wild-type allele is surrounded by a large enough number of drive homozygous individuals, resulting in a very low chance of wild-type recolonisation. To do so, we make a deterministic approximation of the distribution of drive alleles within the wave, and we split the distribution of wild-type alleles into a deterministic part and a stochastic part. Our analytical and numerical results suggest that the probability of wild-type recolonisation events increases with lower fitness of drive individuals, with smaller migration rate, and also with smaller local carrying capacity. Numerical simulations show that these results extend to two spatial dimensions. We also demonstrate that, if a wild-type recolonisation event were to occur, the probability of a following drive reinvasion event decreases with smaller values of the intrinsic growth rate of the population. Overall, our study paves the way for further analysis of wild-type recolonisation at the back of eradication traveling waves.

Hilary Putnam's biography and philosophical development reflect the history of Anglo-Saxon philosophy over the last 40 years. Putnam has influenced this history significantly for almost as long. In this introduction, the main aim is to present the context in which Putnam stands and from which his philosophical contributions can be understood. In the context of a sketch of Putnam's philosophical development, a preliminary historical classification of his work will also be attempted, even if this is not the place for a comprehensive critique or presentation: The introduction must remain at a fairly elementary level and of course cannot replace a reading of the texts. Since Putnam's work is certainly part of a rapprochement between 'analytic' and 'continental' philosophy, the introduction to the texts translated here should finally make clear what Putnam has to offer non-analytically oriented readers. Hilary Putnams Biographie und philosophische Entwicklung spiegeln die Geschichte der angels\"achsischen Philosophie in den letzten 40 Jahren. Beinahe ebenso lange hat Putnam diese Geschichte wesentlich beeinflu{\ss}t. In der vorliegenden Einleitung soll vor allem der Kontext dargestellt werden, in dem Putnam steht und aus dem heraus verst\"andlich wird, was er philosophisch zu sagen hat. Im Rahmen einer Skizze von Putnams philosophischer Entwicklung soll zudem eine vorl\"aufige philosophiehistorische Einordnung versucht werden, auch wenn hier nicht der Ort f\"ur eine umfassende Kritik oder Darstellung sein kann: Die Einleitung mu{\ss} auf recht elementarem Niveau bleiben und kann eine Lekt\"ure der Texte nat\"urlich nicht ersetzen. Da Putnams Werk sicherlich Teil einer Ann\"aherung von 'analytischer' und 'kontinentaler' Philosophie ist, soll bei der Einf\"uhrung in die hier \"ubersetzten Texte schlie{\ss}lich deutlich werden, was Putnam nicht analytisch orientierten Lesern zu bieten hat., Comment: in German language

We prove the modularity of a positive proportion of abelian surfaces over $\mathbf{Q}$. More precisely, we prove the modularity of abelian surfaces which are ordinary at $3$ and are $3$-distinguished, subject to some assumptions on the $3$-torsion representation (a "big image" hypothesis, and a technical hypothesis on the action of a decomposition group at $2$). We employ a 2-3 switch and a new classicality theorem (in the style of Lue Pan) for ordinary $p$-adic Siegel modular forms., Comment: 230 pages, comments welcome

Our approach to training 3D vision-language understanding models is to train a feedforward model that makes predictions in 3D, but never requires 3D labels and is supervised only in 2D, using 2D losses and differentiable rendering. The approach is new for vision-language understanding. By treating the reconstruction as a ``latent variable'', we can render the outputs without placing unnecessary constraints on the network architecture (e.g. can be used with decoder-only models). For training, only need images and camera pose, and 2D labels. We show that we can even remove the need for 2D labels by using pseudo-labels from pretrained 2D models. We demonstrate this to pretrain a network, and we finetune it for 3D vision-language understanding tasks. We show this approach outperforms baselines/sota for 3D vision-language grounding, and also outperforms other 3D pretraining techniques. Project page: https://liftgs.github.io., Comment: Project page: https://liftgs.github.io

Reinforcement learning algorithms typically necessitate extensive exploration of the state space to find optimal policies. However, in safety-critical applications, the risks associated with such exploration can lead to catastrophic consequences. Existing safe exploration methods attempt to mitigate this by imposing constraints, which often result in overly conservative behaviours and inefficient learning. Heavy penalties for early constraint violations can trap agents in local optima, deterring exploration of risky yet high-reward regions of the state space. To address this, we introduce a method that explicitly learns state-conditioned safety representations. By augmenting the state features with these safety representations, our approach naturally encourages safer exploration without being excessively cautious, resulting in more efficient and safer policy learning in safety-critical scenarios. Empirical evaluations across diverse environments show that our method significantly improves task performance while reducing constraint violations during training, underscoring its effectiveness in balancing exploration with safety., Comment: Accepted at International Conference on Learning Representations (ICLR) 2025

Altermagnetism defies conventional classifications of collinear magnetic phases, standing apart from ferromagnetism and antiferromagnetism with its unique combination of spin-dependent symmetries, net-zero magnetization, and anomalous Hall transport. Although altermagnetic states have been realized experimentally, their integration into functional devices has been hindered by the structural rigidity and poor tunability of existing materials. First, through cobalt intercalation of the superconducting 2H-NbSe$_2$ polymorph, we induce and stabilize a robust altermagnetic phase and using both theory and experiment, we directly observe the lifting of Kramers degeneracy. Then, using ultrafast laser pulses, we demonstrate how the low temperature phase of this system can be quenched, realizing the first example of an optical altermagnetic switch. While shedding light on overlooked aspects of altermagnetism, our findings open pathways to spin-based technologies and lay a foundation for advancing the emerging field of altertronics.

The kagome lattice stands as a rich platform for hosting a wide array of correlated quantum phenomena, ranging from charge density waves and superconductivity to electron nematicity and loop current states. Direct detection of loop currents in kagome systems has remained a formidable challenge due to their intricate spatial arrangements and the weak magnetic field signatures they produce. This has left their existence and underlying mechanisms a topic of intense debate. In this work, we uncover a hallmark reconcilable with loop currents: spin handedness-selective signals that surpass conventional dichroic, spin, and spin-dichroic responses. We observe this phenomenon in the kagome metal CsTi$_3$Bi$_5$ and we call it the anomalous spin-optical helical effect. This effect arises from the coupling of light' s helicity with spin-orbital electron correlations, providing a groundbreaking method to visualize loop currents in quantum materials. Our discovery not only enriches the debate surrounding loop currents but also paves the way for new strategies to exploit the electronic phases of quantum materials via light-matter interaction.

Nulling interferometry is a promising technique for direct detection of exoplanets. However, the performance of current devices is limited by different perturbations sources and especially by its sensitivity to any phase aberrations. The work presented here attempts to overcome those limitations by using a four-telescopes nulling interferometer architecture, called Kernel-Nuller, which includes a recombiner that positions the four signals in phase quadrature. This architecture is based on an integrated optical component containing 14 electronically controlled phase shifters, used to correct optical path differences that would be induced by manufacturing defects. The first part of the study consists in the development of an algorithm providing the delays to be injected into the component to optimize the performance of that device. The next step of this study deals with the analysis of the intensity distributions produced at the output of the Kernel-Nuller through a series of observations. Then we apply statistical tests and data treatment techniques to detect the signature of an exoplanets.

The $\Xi_c^+ \pi^- \pi^+$ spectrum is investigated using proton-proton collisions at a center-of-mass energy of 13TeV, corresponding to an integrated luminosity of 5.4fb$^{-1}$, collected by the LHCb experiment during 2016--2018. Four states are observed with high significance, and their masses and widths are measured to be \begin{align*} m[\Xi_c(2815)^{+}] &= 2816.65 \pm 0.03 \pm 0.03 \pm 0.23 ~\text{MeV}, \Gamma[\Xi_c(2815)^{+}] &= 2.07 \pm 0.08 \pm 0.12~\text{MeV},\\[5pt] m[\Xi_c(2923)^{+}] &= 2922.8 \pm 0.3 \pm 0.5 \pm 0.2~\text{MeV}, \Gamma[\Xi_c(2923)^{+}] &= 5.3 \pm 0.9 \pm 1.4~\text{MeV},\\[5pt] m[\Xi_c(2970)^{+}] &= 2968.6 \pm 0.5 \pm 0.5 \pm 0.2~\text{MeV}, \Gamma[\Xi_c(2970)^{+}] &= 31.7 \pm 1.7 \pm 1.9~\text{MeV},\\[5pt] m[\Xi_c(3080)^{+}] &= 3076.8 \pm 0.7 \pm 1.3 \pm 0.2~\text{MeV}, \Gamma[\Xi_c(3080)^{+}] &= 6.8 \pm 2.3 \pm 0.9~\text{MeV}, \end{align*} where the uncertainties are statistical, systematic, and due to the limited precision on the $\Xi_c^+$ mass, respectively. The $\Xi_c(2923)^{+}$ baryon is observed for the first time, and is consistent with being the isospin partner of the previously observed $\Xi_c(2923)^{0}$ state. Most of the measured parameters are more precise than existing world averages., Comment: All figures and tables, along with machine-readable versions and any supplementary material and additional information, are available at https://lbfence.cern.ch/alcm/public/analysis/full-details/3080/ (LHCb public pages)

Frequency Phase Transfer (FPT) is a technique designed to increase coherence and sensitivity in radio interferometry by making use of the non-dispersive nature of the troposphere to calibrate high-frequency data using solutions derived at a lower frequency. While the Korean VLBI Network has pioneered the use of simultaneous multi-band systems for routine FPT up to an observing frequency of 130 GHz, this technique remains largely untested in the (sub)millimeter regime. A recent effort has been made to outfit dual-band systems at (sub)millimeter observatories participating in the Event Horizon Telescope (EHT) and to test the feasibility and performance of FPT up to the observing frequencies of the EHT. We present the results of simultaneous dual-frequency observations conducted in January 2024 on an Earth-sized baseline between the IRAM 30-m in Spain and the JCMT and SMA in Hawai`i. We performed simultaneous observations at 86 and 215 GHz on the bright sources J0958+6533 and OJ287, with strong detections obtained at both frequencies. We observe a strong correlation between the interferometric phases at the two frequencies, matching the trend expected for atmospheric fluctuations and demonstrating for the first time the viability of FPT for VLBI at a wavelength of $\sim$1 millimeter. We show that the application of FPT systematically increases the 215 GHz coherence on all averaging timescales. In addition, the use of the co-located JCMT and SMA as a single dual-frequency station demonstrates the feasibility of paired-antenna FPT for VLBI for the first time, with implications for future array capabilities (e.g., ALMA sub-arraying and ngVLA calibration strategies)., Comment: 11 pages, 5 figures, accepted to AJ

Variational inequalities (VIs) encompass many fundamental problems in diverse areas ranging from engineering to economics and machine learning. However, their considerable expressivity comes at the cost of computational intractability. In this paper, we introduce and analyze a natural relaxation -- which we refer to as expected variational inequalities (EVIs) -- where the goal is to find a distribution that satisfies the VI constraint in expectation. By adapting recent techniques from game theory, we show that, unlike VIs, EVIs can be solved in polynomial time under general (nonmonotone) operators. EVIs capture the seminal notion of correlated equilibria, but enjoy a greater reach beyond games. We also employ our framework to capture and generalize several existing disparate results, including from settings such as smooth games, and games with coupled constraints or nonconcave utilities., Comment: V2 expands on the related work

$\Phi$-equilibria -- and the associated notion of $\Phi$-regret -- are a powerful and flexible framework at the heart of online learning and game theory, whereby enriching the set of deviations $\Phi$ begets stronger notions of rationality. Recently, Daskalakis, Farina, Fishelson, Pipis, and Schneider (STOC '24) -- abbreviated as DFFPS -- settled the existence of efficient algorithms when $\Phi$ contains only linear maps under a general, $d$-dimensional convex constraint set $\mathcal{X}$. In this paper, we significantly extend their work by resolving the case where $\Phi$ is $k$-dimensional; degree-$\ell$ polynomials constitute a canonical such example with $k = d^{O(\ell)}$. In particular, positing only oracle access to $\mathcal{X}$, we obtain two main positive results: i) a $\text{poly}(n, d, k, \text{log}(1/\epsilon))$-time algorithm for computing $\epsilon$-approximate $\Phi$-equilibria in $n$-player multilinear games, and ii) an efficient online algorithm that incurs average $\Phi$-regret at most $\epsilon$ using $\text{poly}(d, k)/\epsilon^2$ rounds. We also show nearly matching lower bounds in the online learning setting, thereby obtaining for the first time a family of deviations that captures the learnability of $\Phi$-regret. From a technical standpoint, we extend the framework of DFFPS from linear maps to the more challenging case of maps with polynomial dimension. At the heart of our approach is a polynomial-time algorithm for computing an expected fixed point of any $\phi : \mathcal{X} \to \mathcal{X}$ based on the ellipsoid against hope (EAH) algorithm of Papadimitriou and Roughgarden (JACM '08). In particular, our algorithm for computing $\Phi$-equilibria is based on executing EAH in a nested fashion -- each step of EAH itself being implemented by invoking a separate call to EAH.

Counterfactual explanations provide actionable insights to achieve desired outcomes by suggesting minimal changes to input features. However, existing methods rely on fixed sets of mutable features, which makes counterfactual explanations inflexible for users with heterogeneous real-world constraints. Here, we introduce Flexible Counterfactual Explanations, a framework incorporating counterfactual templates, which allows users to dynamically specify mutable features at inference time. In our implementation, we use Generative Adversarial Networks (FCEGAN), which align explanations with user-defined constraints without requiring model retraining or additional optimization. Furthermore, FCEGAN is designed for black-box scenarios, leveraging historical prediction datasets to generate explanations without direct access to model internals. Experiments across economic and healthcare datasets demonstrate that FCEGAN significantly improves counterfactual explanations' validity compared to traditional benchmark methods. By integrating user-driven flexibility and black-box compatibility, counterfactual templates support personalized explanations tailored to user constraints., Comment: 28 pages, 13 figures

In Federated Learning (FL), several clients jointly learn a machine learning model: each client maintains a local model for its local learning dataset, while a master server maintains a global model by aggregating the local models of the client devices. However, the repetitive communication between server and clients leaves room for attacks aimed at compromising the integrity of the global model, causing errors in its targeted predictions. In response to such threats on FL, various defense measures have been proposed in the literature. In this paper, we present a powerful defense against malicious clients in FL, called FedSV, using the Shapley Value (SV), which has been proposed recently to measure user contribution in FL by computing the marginal increase of average accuracy of the model due to the addition of local data of a user. Our approach makes the identification of malicious clients more robust, since during the learning phase, it estimates the contribution of each client according to the different groups to which the target client belongs. FedSV's effectiveness is demonstrated by extensive experiments on MNIST datasets in a cross-silo context under various attacks.

The safety alignment of large language models (LLMs) can be circumvented through adversarially crafted inputs, yet the mechanisms by which these attacks bypass safety barriers remain poorly understood. Prior work suggests that a single refusal direction in the model's activation space determines whether an LLM refuses a request. In this study, we propose a novel gradient-based approach to representation engineering and use it to identify refusal directions. Contrary to prior work, we uncover multiple independent directions and even multi-dimensional concept cones that mediate refusal. Moreover, we show that orthogonality alone does not imply independence under intervention, motivating the notion of representational independence that accounts for both linear and non-linear effects. Using this framework, we identify mechanistically independent refusal directions. We show that refusal mechanisms in LLMs are governed by complex spatial structures and identify functionally independent directions, confirming that multiple distinct mechanisms drive refusal behavior. Our gradient-based approach uncovers these mechanisms and can further serve as a foundation for future work on understanding LLMs.

To circumvent the alignment of large language models (LLMs), current optimization-based adversarial attacks usually craft adversarial prompts by maximizing the likelihood of a so-called affirmative response. An affirmative response is a manually designed start of a harmful answer to an inappropriate request. While it is often easy to craft prompts that yield a substantial likelihood for the affirmative response, the attacked model frequently does not complete the response in a harmful manner. Moreover, the affirmative objective is usually not adapted to model-specific preferences and essentially ignores the fact that LLMs output a distribution over responses. If low attack success under such an objective is taken as a measure of robustness, the true robustness might be grossly overestimated. To alleviate these flaws, we propose an adaptive and semantic optimization problem over the population of responses. We derive a generally applicable objective via the REINFORCE policy-gradient formalism and demonstrate its efficacy with the state-of-the-art jailbreak algorithms Greedy Coordinate Gradient (GCG) and Projected Gradient Descent (PGD). For example, our objective doubles the attack success rate (ASR) on Llama3 and increases the ASR from 2% to 50% with circuit breaker defense., Comment: 30 pages, 6 figures, 15 tables

We study two axioms for social choice functions that capture the impact of similar candidates: independence of clones (IoC) and composition consistency (CC). We clarify the relationship between these axioms by observing that CC is strictly more demanding than IoC, and investigate whether common voting rules that are known to be independent of clones (such as STV, Ranked Pairs, Schulze, and Split Cycle) are composition-consistent. While for most of these rules the answer is negative, we identify a variant of Ranked Pairs that satisfies CC. Further, we show how to efficiently modify any (neutral) social choice function so that it satisfies CC, while maintaining its other desirable properties. Our transformation relies on the hierarchical representation of clone structures via PQ-trees. We extend our analysis to social preference functions. Finally, we interpret IoC and CC as measures of robustness against strategic manipulation by candidates, with IoC corresponding to strategy-proofness and CC corresponding to obvious strategy-proofness., Comment: 60 pages, 9 figures

First-order logic (FOL) can represent the logical entailment semantics of natural language (NL) sentences, but determining natural language entailment using FOL remains a challenge. To address this, we propose the Entailment-Preserving FOL representations (EPF) task and introduce reference-free evaluation metrics for EPF, the Entailment-Preserving Rate (EPR) family. In EPF, one should generate FOL representations from multi-premise natural language entailment data (e.g. EntailmentBank) so that the automatic prover's result preserves the entailment labels. Experiments show that existing methods for NL-to-FOL translation struggle in EPF. To this extent, we propose a training method specialized for the task, iterative learning-to-rank, which directly optimizes the model's EPR score through a novel scoring function and a learning-to-rank objective. Our method achieves a 1.8-2.7% improvement in EPR and a 17.4-20.6% increase in EPR@16 compared to diverse baselines in three datasets. Further analyses reveal that iterative learning-to-rank effectively suppresses the arbitrariness of FOL representation by reducing the diversity of predicate signatures, and maintains strong performance across diverse inference types and out-of-domain data., Comment: 14 pages (8 pages of main content), 8 figures

Quantum compilers play a crucial role in quantum computing by converting these algorithmic quantum circuits into forms compatible with specific quantum computer hardware. However, untrusted quantum compilers present considerable risks, including the potential theft of quantum circuit intellectual property (IP) and compromise of the functionality (e.g. Trojan insertion). Quantum circuit obfuscation techniques protect quantum IP by transforming a quantum circuit into a key-dependent version before compilation and restoring the compiled circuit's functionality with the correct key. This prevents the untrusted compiler from knowing the circuit's original functionality. Existing quantum circuit obfuscation techniques focus on inserting key qubits to control key gates. One added key gate can represent at most one Boolean key bit. In this paper, we propose OPAQUE, a phase-based quantum circuit obfuscation approach where we use the angle of rotation gates as the secret keys. The rotation angle is a continuous value, which makes it possible to represent multiple key bits. Moreover, phase gates are usually implemented as virtual gates in quantum hardware, diminishing their cost and impact on accuracy., Comment: To appear at ISQED '25

A flexible intelligent metasurface (FIM) is composed of an array of low-cost radiating elements, each of which can independently radiate electromagnetic signals and flexibly adjust its position through a 3D surface-morphing process. In our system, an FIM is deployed at a base station (BS) that transmits to multiple single-antenna users. We formulate an optimization problem for minimizing the total downlink transmit power at the BS by jointly optimizing the transmit beamforming and the FIM's surface shape, subject to an individual signal-to-interference-plus-noise ratio (SINR) constraint for each user as well as to a constraint on the maximum morphing range of the FIM. To address this problem, an efficient alternating optimization method is proposed to iteratively update the FIM's surface shape and the transmit beamformer to gradually reduce the transmit power. Finally, our simulation results show that at a given data rate the FIM reduces the transmit power by about $3$ dB compared to conventional rigid 2D arrays., Comment: 6 pages, 4 figures, presented at IEEE GLOBECOM 2024

We present a new method for generating confidence sets within the split conformal prediction framework. Our method performs a trainable transformation of any given conformity score to improve conditional coverage while ensuring exact marginal coverage. The transformation is based on an estimate of the conditional quantile of conformity scores. The resulting method is particularly beneficial for constructing adaptive confidence sets in multi-output problems where standard conformal quantile regression approaches have limited applicability. We develop a theoretical bound that captures the influence of the accuracy of the quantile estimate on the approximate conditional validity, unlike classical bounds for conformal prediction methods that only offer marginal coverage. We experimentally show that our method is highly adaptive to the local data structure and outperforms existing methods in terms of conditional coverage, improving the reliability of statistical inference in various applications.