Search

Large language models (LLMs) are vulnerable to universal jailbreaks-prompting strategies that systematically bypass model safeguards and enable users to carry out harmful processes that require many model interactions, like manufacturing illegal substances at scale. To defend against these attacks, we introduce Constitutional Classifiers: safeguards trained on synthetic data, generated by prompting LLMs with natural language rules (i.e., a constitution) specifying permitted and restricted content. In over 3,000 estimated hours of red teaming, no red teamer found a universal jailbreak that could extract information from an early classifier-guarded LLM at a similar level of detail to an unguarded model across most target queries. On automated evaluations, enhanced classifiers demonstrated robust defense against held-out domain-specific jailbreaks. These classifiers also maintain deployment viability, with an absolute 0.38% increase in production-traffic refusals and a 23.7% inference overhead. Our work demonstrates that defending against universal jailbreaks while maintaining practical deployment viability is tractable.

We report the first detections of [Ne V] 14.3 {\mu}m and [Ne VI] 7.7 {\mu}m at high confidence (S/N>=6) in the nuclear region of the nearby spiral galaxy M83. Emission line maps of these high ionization lines show several compact structures. Specifically, the [Ne VI] emission is located at 140 pc from the optical nucleus and appears as a point source of size ~<18 pc (FWHM =<0.8"). We investigate the possible source of this extreme emission through comparison with photoionization models and ancillary data. We find that photoionization models of fast radiative shocks are able to reproduce the observed high excitation emission line fluxes only for the lowest preshock density available in the library, n =0.01 cm^-3. Additionally, tailored active galactic nuclei (AGN) photoionization models assuming a two-zone structure are compatible with the observed high ionization fluxes. Our simple AGN model shows that the emission at the location of the [Ne VI] source can be the result of a cloud being ionized by the radiation cone of an AGN. We stress, however, that to definitively confirm an AGN as the main source of the observed emission, more complex modeling accounting for different geometries is required. Previously known as a purely starburst system, these new findings of the nuclear region of M83 will require a reassessment of its nature and of objects similar to it, particularly now that we have access to the unparalleled infrared sensitivity and spatial resolution of the James Webb Space Telescope., Comment: 12 pages, 8 figures, 1 table, Accepted for publication in ApJ

A new model for domination reconfiguration is introduced which combines the properties of the preexisting token addition/removal (TAR) and token sliding (TS) models. The vertices of the TARS-graph correspond to the dominating sets of $G$, where two vertices are adjacent if and only if they are adjacent via either the TAR reconfiguration rule or the TS reconfiguration rule. While the domination reconfiguration graph obtained by using only the TAR rule (sometimes called the dominating graph) will never have a Hamilton cycle, we show that for some classes of graphs $G$, by adding a relatively small number of token sliding edges, the resulting graph is not only hamiltonian, but is in fact pancyclic. In particular, if the underlying graphs are trees, complete graphs, or complete multipartite graphs, their TARS-graphs will be pancyclic. We also provide pancyclicity results for TARS-graphs of graph unions and joins, and conclude by posing the question: Are all TARS-graphs pancyclic?, Comment: 14 pages, 4 figures

We present 3 years of high-contrast imaging of the PDS 70 b and c accreting protoplanets with the new extreme AO system MagAO-X as part of the MaxProtoPlanetS survey of H$\alpha$ protoplanets. In 2023 and 2024 our sharp (25-27 mas FWHM); well AO corrected (20-26% Strehl), deep (2-3.6hr) images detect compact (r~30 mas; r~3 au) circumplanetary disks (CPDs) surrounding both protoplanets. Starlight scattering off the dusty outer edges of these CPDs is the likely source of the bright compact continuum light detected within ~30 mas of both planets in our simultaneously obtained continuum 668 nm filter images. After subtraction of contaminating continuum and PSF residuals with pyKLIP ADI and SDI we obtained high-contrast ASDI H$\alpha$ images of both planets in 2022, 2023 and 2024. We find the H$\alpha$ line flux of planet b fell by (8.1$\pm$1.6)x10$^{-16}$ ergs/s/cm$^2$ a factor of 4.6 drop in flux from 2022 to 2023. In March 2024, planet b continued to be faint with just a slight 1.6x rise to an H$\alpha$ line flux of (3.64$\pm$0.87)x10$^{-16}$ ergs/s/cm$^2$. For c we measure a significant increase of (2.74$\pm$0.51)x10$^{-16}$ ergs/s/cm$^2$ from 2023 to 2024 which is a factor of 2.3x increase. So both protoplanets have recently experienced significant H$\alpha$ variability with ~1 yr sampling. In 2024, planet c is brighter than b: as c is brightening and b generally fading. We also tentatively detect one new point source "CC3" inside the inner disk (~49 mas; at PA~295 deg; 2024) with orbital motion roughly consistent with a ~5.6 au orbit., Comment: 50 pages, 20 figures, published in January 2025 Astronomical Journal. NOTE Figure D1 has correct planetary mass accretion scale compared to published AJ version

We present an end-to-end framework for generating synthetic users for evaluating interactive agents designed to encourage positive behavior changes, such as in health and lifestyle coaching. The synthetic users are grounded in health and lifestyle conditions, specifically sleep and diabetes management in this study, to ensure realistic interactions with the health coaching agent. Synthetic users are created in two stages: first, structured data are generated grounded in real-world health and lifestyle factors in addition to basic demographics and behavioral attributes; second, full profiles of the synthetic users are developed conditioned on the structured data. Interactions between synthetic users and the coaching agent are simulated using generative agent-based models such as Concordia, or directly by prompting a language model. Using two independently-developed agents for sleep and diabetes coaching as case studies, the validity of this framework is demonstrated by analyzing the coaching agent's understanding of the synthetic users' needs and challenges. Finally, through multiple blinded evaluations of user-coach interactions by human experts, we demonstrate that our synthetic users with health and behavioral attributes more accurately portray real human users with the same attributes, compared to generic synthetic users not grounded in such attributes. The proposed framework lays the foundation for efficient development of conversational agents through extensive, realistic, and grounded simulated interactions.

Coupling excitons with quantized radiation has been shown to enable coherent ballistic transport at room temperature inside optical cavities. Previous theoretical works employ a simple description of the material, depicting it as a single layer placed in the middle of the optical cavity, thereby ignoring the spatial variation of the radiation in the cavity quantization direction. In contrast, in most experiments, the optical cavity is filled with organic molecules or multilayered materials. Here, we develop an efficient mixed-quantum-classical approach, introducing a {\it bright layer} description, to simulate the exciton-polariton quantum dynamics. Our simulations reveal that, for the same Rabi splitting, a multilayer material extends the quantum coherence lifetime and enhances transport compared to a single-layer material. We find that this enhanced coherence can be traced to a synchronization of phonon fluctuations over multiple layers, wherein the collective light-matter coupling in a multilayered material effectively suppresses the phonon-induced dynamical disorder.

Ballistic heat transport and second sound propagation in solids is of direct relevance in electronic and energy applications at short length scales and low temperatures. Measurement or calculation of thermal conductivity, which is typically a primary objective, may be of limited usefulness for predicting heat transport which does not follow the heat-diffusion equation. In this paper, molecular-dynamics simulations of hexagonal BN (h-BN) are used to compute thermal response functions from equilibrium correlation functions defined in Fourier space. The response functions are useful for describing the time-dependent transport beyond the usual assumptions of Fourier's law. The results demonstrate that for length scales ~110nm at T=100K second sound should be experimentally observable. At higher temperatures and longer length scales, while second sound may not be directly observable, thermal transport can nevertheless strongly deviate from predictions based on the heat-diffusion equation. Along with classical simulations, we outline a first-principles, many-body theoretical approach for calculation of the response function based on solutions of the Bethe-Salpeter equation. The relevant expressions for heat current clarify the importance of phase coherence within a phonon branch to the observation of second sound. Previous work on one-dimensional chains is also discussed to show that materials characterized by linear dispersion and simple phonon band structure should more readily display second sound., Comment: 27 pages, 11 figures, to be submitted to Phys. Rev. B

Nearby blue compact dwarf galaxies (BCDs) share similar properties with objects from the Epoch of Reionization revealed by JWST, in terms of low stellar mass, low metallicity and high specific star-formation rate. Thus, they represent ideal local laboratories for detailed multi-wavelength studies to understand their properties and the mechanisms shaping them. We report the first JWST MIRI/MRS observations of the BCD SBS 0335-052 E, analyzing MIR emission lines tracing different levels of ionization (e.g., [NeII], [SIV], [NeIII], [OIV], [NeV]) of the ionized gas. SBS 0335-052 E MIR emission is characterized by a bright point source, located in one of the youngest and most embedded stellar clusters ($t\sim3$ Myr, $A_V\sim20$), and underlying extended high-ionization emission (i.e., [OIV], [NeV]) from the surroundings of the older and less dusty stellar clusters ($t< 20 $ Myr, $A_V\sim8$). From the comparison with state-of-the-art models, we can exclude shocks, X-ray binaries, and old stellar populations as the main sources of the high ionization. Interestingly, a 4-16% contribution of a $\sim10^5$ M$_\odot$ intermediate massive black hole (IMBH) is needed to justify the strong [NeV]/[NeII] and would be consistent with optical/UV line ratios from previous studies. However, even IMBH models cannot explain the strongest [OIV]/[NeIII]. Also, star-forming models (regardless of including X-ray binaries) struggle to reproduce even the lower ionization line ratios (e.g., [SIV]/[NeII]) typically observed in BCDs. Overall, while current models suggest the need to account for an accreting IMBH in this high-$z$ analog, limitations still exist in predicting high-ionization emission lines (I.P. $>54$ eV) when modeling these low-metallicity environments, thus other sources of ionization cannot be fully ruled out., Comment: Submitted to ApJ

Traditional data storage formats and databases often introduce complexities and inefficiencies that hinder rapid iteration and adaptability. To address these challenges, we introduce ParquetDB, a Python-based database framework that leverages the Parquet file format's optimized columnar storage. ParquetDB offers efficient serialization and deserialization, native support for complex and nested data types, reduced dependency on indexing through predicate pushdown filtering, and enhanced portability due to its file-based storage system. Benchmarks show that ParquetDB outperforms traditional databases like SQLite and MongoDB in managing large volumes of data, especially when using data formats compatible with PyArrow. We validate ParquetDB's practical utility by applying it to the Alexandria 3D Materials Database, efficiently handling approximately 4.8 million complex and nested records. By addressing the inherent limitations of existing data storage systems and continuously evolving to meet future demands, ParquetDB has the potential to significantly streamline data management processes and accelerate research development in data-driven fields., Comment: 41 pages, 12 figures

Two-stage exams, which pair a traditional individual exam with a subsequent collaborative exam, are widely popular with both students and faculty for fostering deep engagement, collaboration, and immediate feedback. Over the last decade, this assessment model has gained substantial traction in STEM college courses; however, holding both stages during class sessions often limits the full potential of two-stage exams, deterring many instructors from adopting them. To accommodate these constraints, instructors must either reduce the exam's content depth and coverage, or hold evening, in-person exam sessions at fixed times, which can often be impractical for students and faculty with external commitments. In this paper, we introduce an alternative asynchronous approach that allows the collaborative portion to be completed unsupervised and outside of regular class hours, within 48 hours of the individual exam. This method not only eases logistical constraints but also allows more time for collaboration, potentially enhancing feedback quality. Our findings show that students equate their engagement, collaboration, and feedback quality during the asynchronous collaborative exam with that of standard in-class exams and show signs of retention of learning. This cost-effective approach requires no extra time or resources and could promote widespread adoption of this effective assessment method, especially in online-only and general STEM courses., Comment: 5 figure, 1 table

Augmented Reality (AR) collaboration can benefit from a shared 2D surface, such as a whiteboard. However, many features of each collaborators physical environment must be considered in order to determine the best placement and shape of the shared surface. We explored the effects of three methods for beginning a collaborative whiteboarding session with varying levels of user control: MANUAL, DISCRETE CHOICE, and AUTOMATIC by conducting a simulated AR study within Virtual Reality (VR). In the MANUAL method, users draw their own surfaces directly in the environment until they agree on the placement; in the DISCRETE CHOICE method, the system provides three options for whiteboard size and location; and in the AUTOMATIC method, the system automatically creates a whiteboard that fits within each collaborators environment. We evaluate these three conditions in a study in which two collaborators used each method to begin collaboration sessions. After establishing a session, the users worked together to complete an affinity diagramming task using the shared whiteboard. We found that the majority of participants preferred to have direct control during the initialization of a new collaboration session, despite the additional workload induced by the Manual method.

A numeration system encodes abstract numeric quantities as concrete strings of written characters. The numeration systems used by modern scripts tend to be precise and unambiguous, but this was not so for the ancient and partially-deciphered proto-Elamite (PE) script, where written numerals can have up to four distinct readings depending on the system that is used to read them. We consider the task of disambiguating between these readings in order to determine the values of the numeric quantities recorded in this corpus. We algorithmically extract a list of possible readings for each PE numeral notation, and contribute two disambiguation techniques based on structural properties of the original documents and classifiers learned with the bootstrapping algorithm. We also contribute a test set for evaluating disambiguation techniques, as well as a novel approach to cautious rule selection for bootstrapped classifiers. Our analysis confirms existing intuitions about this script and reveals previously-unknown correlations between tablet content and numeral magnitude. This work is crucial to understanding and deciphering PE, as the corpus is heavily accounting-focused and contains many more numeric tokens than tokens of text.

We present progress on the calculation of scalar, vector, and tensor form factors for the following meson decays: $B\rightarrow\pi$, $B_s\rightarrow K$, $D\rightarrow\pi$ and $D_s\rightarrow K$. This calculation uses the MILC HISQ gluon field ensembles with HISQ valence quarks. We generate ensembles of correlator data with varying lattice spacings, some as small as 0.044 fm. Some ensembles have a strange-to-light quark mass ratio of 5:1 and others use the physical light quark mass. The fully-relativistic, heavy-HISQ approach is used for the heavy quark, with simulation masses ranging from the charm to near the bottom. This heavy-HISQ approach provides nearly full coverage of the kinematic range., Comment: 10 pages, 5 figures. Presentation on this work was given at the 41st International Symposium on Lattice Field Theory 28 July - 3 August 2024, Liverpool, UK. Submission ID: PoS(LATTICE2024)262

We develop an analytical microscopic theory to describe the polaron-polariton dispersion, formed by hybridizing excitons, photons, and phonons, and their coherent dynamics inside optical cavities. Starting from a microscopic light-matter Hamiltonian, we derive a simple analytical model by pursuing a non-perturbative treatment of the phonon and photon couplings to excitons. Within our theoretical framework, the phonons are treated as classical fields that are then quantized via the Floquet formalism. We show that, to a good approximation, the entire polaron-polariton system can be described using a band picture despite the phonons breaking translational symmetry. Our theory also sheds light on the long-lived coherent ballistic motion of exciton-polaritons with high excitonic character that propagate with group velocities lower than is expected from pure exciton-polariton bands, offering a microscopic explanation for these puzzling experimental observations.

Current AI frameworks for brain decoding and encoding, typically train and test models within the same datasets. This limits their utility for brain computer interfaces (BCI) or neurofeedback, for which it would be useful to pool experiences across individuals to better simulate stimuli not sampled during training. A key obstacle to model generalisation is the degree of variability of inter-subject cortical organisation, which makes it difficult to align or compare cortical signals across participants. In this paper we address this through the use of surface vision transformers, which build a generalisable model of cortical functional dynamics, through encoding the topography of cortical networks and their interactions as a moving image across a surface. This is then combined with tri-modal self-supervised contrastive (CLIP) alignment of audio, video, and fMRI modalities to enable the retrieval of visual and auditory stimuli from patterns of cortical activity (and vice-versa). We validate our approach on 7T task-fMRI data from 174 healthy participants engaged in the movie-watching experiment from the Human Connectome Project (HCP). Results show that it is possible to detect which movie clips an individual is watching purely from their brain activity, even for individuals and movies not seen during training. Further analysis of attention maps reveals that our model captures individual patterns of brain activity that reflect semantic and visual systems. This opens the door to future personalised simulations of brain function. Code & pre-trained models will be made available at https://github.com/metrics-lab/sim, processed data for training will be available upon request at https://gin.g-node.org/Sdahan30/sim., Comment: 27 pages, accepted to ICLR 2025

Following a method introduced by Thomas-Vasquez and developed by Grundman, we prove that many Hilbert modular threefolds of arithmetic genus $0$ and $1$ are of general type, and that some are of nonnegative Kodaira dimension. The new ingredient is a detailed study of the geometry and combinatorics of totally positive integral elements $x$ of a fractional ideal $I$ in a totally real number field $K$ with the property that $\mathop{\mathrm{tr}} xy < \mathop{\mathrm{min}} I \mathop{\mathrm{tr}} y$ for some $y \gg 0 \in K$., Comment: 25 pages. Submitted to LUCANT 2025 conference. Comments welcome

Distracted driving remains a significant global challenge with severe human and economic repercussions, demanding improved detection and intervention strategies. While previous studies have extensively explored single-modality approaches, recent research indicates that these systems often fall short in identifying complex distraction patterns, particularly cognitive distractions. This systematic review addresses critical gaps by providing a comprehensive analysis of machine learning (ML) and deep learning (DL) techniques applied across various data modalities - visual,, sensory, auditory, and multimodal. By categorizing and evaluating studies based on modality, data accessibility, and methodology, this review clarifies which approaches yield the highest accuracy and are best suited for specific distracted driving detection goals. The findings offer clear guidance on the advantages of multimodal versus single-modal systems and capture the latest advancements in the field. Ultimately, this review contributes valuable insights for developing robust distracted driving detection frameworks, supporting enhanced road safety and mitigation strategies.

We present MOFA, an open-source generative AI (GenAI) plus simulation workflow for high-throughput generation of metal-organic frameworks (MOFs) on large-scale high-performance computing (HPC) systems. MOFA addresses key challenges in integrating GPU-accelerated computing for GPU-intensive GenAI tasks, including distributed training and inference, alongside CPU- and GPU-optimized tasks for screening and filtering AI-generated MOFs using molecular dynamics, density functional theory, and Monte Carlo simulations. These heterogeneous tasks are unified within an online learning framework that optimizes the utilization of available CPU and GPU resources across HPC systems. Performance metrics from a 450-node (14,400 AMD Zen 3 CPUs + 1800 NVIDIA A100 GPUs) supercomputer run demonstrate that MOFA achieves high-throughput generation of novel MOF structures, with CO$_2$ adsorption capacities ranking among the top 10 in the hypothetical MOF (hMOF) dataset. Furthermore, the production of high-quality MOFs exhibits a linear relationship with the number of nodes utilized. The modular architecture of MOFA will facilitate its integration into other scientific applications that dynamically combine GenAI with large-scale simulations., Comment: 13 pages, 10 figures

Structural Equation Models (SEM) are the standard approach to representing causal dependencies between variables in causal models. In this paper we propose a new interpretation of SEMs when reasoning about Actual Causality, in which SEMs are viewed as mechanisms transforming the dynamics of exogenous variables into the dynamics of endogenous variables. This allows us to combine counterfactual causal reasoning with existing temporal logic formalisms, and to introduce a temporal logic, CPLTL, for causal reasoning about such structures. We show that the standard restriction to so-called \textit{recursive} models (with no cycles in the dependency graph) is not necessary in our approach, allowing us to reason about mutually dependent processes and feedback loops. Finally, we introduce new notions of model equivalence for temporal causal models, and show that CPLTL has an efficient model-checking procedure.

This roadmap consolidates recent advances while exploring emerging applications, reflecting the remarkable diversity of hardware platforms, neuromorphic concepts, and implementation philosophies reported in the field. It emphasizes the critical role of cross-disciplinary collaboration in this rapidly evolving field.

The large apertures of the upcoming generation of Giant Segmented Mirror Telescopes will enable unprecedented angular resolutions that scale as $\propto$ $\lambda$/D and higher sensitivities that scale as $D^4$ for point sources corrected by adaptive optics. However, all will have pupil segmentation caused by mechanical struts holding up the secondary mirror [European Extremely Large Telescope and Thirty Meter Telescope] or intrinsically, by design, as in the Giant Magellan Telescope. These gaps will be separated by more than a typical atmospheric coherence length (Fried Parameter). The pupil fragmentation at scales larger than the typical atmospheric coherence length, combined with wavefront sensors with weak or ambiguous sensitivity to differential piston, can introduce differential piston areas of the wavefront known as "petal modes". Commonly used wavefront sensors, such as a pyramid WFS, also struggle with phase wrapping caused by >$\lambda$/2 differential piston WFE. We have developed the holographic dispersed fringe sensor, a single pupil-plane optic that employs holography to interfere the dispersed light from each segment onto different spatial locations in the focal plane to sense and correct differential piston between the segments. This allows for a very high and linear dynamic piston sensing range of approximately $\pm$10 $\mu$m. We have begun the initial attempts at phasing a segmented pupil utilizing the HDFS on the High Contrast Adaptive optics phasing Testbed and the Extreme Magellan Adaptive Optics instrument (MagAO-X) at the University of Arizona. Additionally, we have demonstrated use of the HDFS as a differential piston sensor on-sky for the first time. We were able to phase each segment to within $\pm\lambda$/11.3 residual piston WFE ($\lambda$ = 800 nm) of a reference segment and achieved ~50 nm RMS residual piston WFE across the aperture in poor seeing conditions., Comment: 34 pages, 26 figures

Transductive zero-shot learning with vision-language models leverages image-image similarities within the dataset to achieve better classification accuracy compared to the inductive setting. However, there is little work that explores the structure of the language space in this context. We propose GTA-CLIP, a novel technique that incorporates supervision from language models for joint transduction in language and vision spaces. Our approach is iterative and consists of three steps: (i) incrementally exploring the attribute space by querying language models, (ii) an attribute-augmented transductive inference procedure, and (iii) fine-tuning the language and vision encoders based on inferred labels within the dataset. Through experiments with CLIP encoders, we demonstrate that GTA-CLIP, yields an average performance improvement of 8.6% and 3.7% across 12 datasets and 3 encoders, over CLIP and transductive CLIP respectively in the zero-shot setting. We also observe similar improvements in a few-shot setting. We present ablation studies that demonstrate the value of each step and visualize how the vision and language spaces evolve over iterations driven by the transductive learning., Comment: Code will be released at https://github.com/cvl-umass/GTA-CLIP

Proteins are composed of linear chains of amino acids that fold into complex three-dimensional structures. A general feature of folded proteins is that they are compact with a radius of gyration $R_g(N) \sim N^{\nu}$ that obeys power-law scaling with the number of amino acids $N$ in each protein and $\nu \sim 1/3$. In this study, we investigate the {\it internal} scaling of the radius of gyration $R_g(n)$ versus the chemical separation $n$ between amino acids for all subchains of length $n$. We show that for globular proteins $R_g(n) \sim n^{\nu_{1,2}}$ with a larger exponent $\nu_1 > 1/3$ for small $n$ and a smaller exponent $\nu_{2} < 1/3$ for large $n$, such that $R_g(N) \propto N^\nu$. To describe this scaling behavior for $R_g(n)$, we carry out folding simulations for a series of coarse-grained models for proteins beginning with the freely-jointed and freely-rotating chain models composed of spherical monomers and varying degrees of stereochemical constraints. We show that a minimal model, which coarse-grains amino acids into a single spherical backbone atom and one variable-sized spherical side-chain atom, and enforces bend- and dihedral-angle constraints, can recapitulate $R_g(n)$ for x-ray crystal structures of globular proteins. In addition, this model predicts the correct average packing fraction and size of the hydrophobic core, which are two key physical features that can be used to distinguish between computational `decoys' and correctly folded proteins in protein design applications., Comment: 8 pages, 6 figures

In this article we propose a game-theoretic approach to the multi-robot task allocation problem using the framework of global games. Each task is associated with a global signal, a real-valued number that captures the task execution progress and/or urgency. We propose a linear objective function for each robot in the system, which, for each task, increases with global signal and decreases with the number assigned robots. We provide conditions on the objective function hyperparameters to induce a mixed Nash equilibrium, i.e., solutions where all robots are not assigned to a single task. The resulting algorithm only requires the inversion of a matrix to determine a probability distribution over the robot assignments. We demonstrate the performance of our algorithm in simulation and provide direction for applications and future work., Comment: 12 pages, 8 figures

Type Ia supernovae are triggered by accretion onto a white dwarf from a companion which is most likely Roche lobe-filling at the time of the explosion. The collision between the ejecta and a surviving companion carves out a conical wake, which could manifest as an asymmetry when the ejecta reaches the remnant phase. We simulate the companion interaction using the Athena++ hydrodynamics solver to determine the ejecta structure for a double-degenerate type Ia supernova. Ejecta in the wake is of lower density and higher velocity than the unperturbed ejecta. We then evolve the ejecta for several thousand years using the expanding-grid code Sprout. The forward shock within the wake is initially indented, but becomes spherical after roughly a thousand years due to transverse motion of shocked ejecta that fills the wake. The reverse shock travels quickly within the wake, leading to an off-center convergence of the reverse shock and leaving the remnant with an asymmetrical core. This also draws material from the interstellar medium deep into the remnant, eventually reaching the center. Large Rayleigh-Taylor plumes are found around the edge of the wake, creating a toroidal structure composed primarily of ejecta. Estimates of the thermal X-ray emission show that such remnants exhibit observable asymmetries for thousands of years., Comment: 17 pages, 14 figures. Submitted to ApJ. Comments welcome!

There has been considerable work on reasoning about the strategic ability of agents under imperfect information. However, existing logics such as Probabilistic Strategy Logic are unable to express properties relating to information transparency. Information transparency concerns the extent to which agents' actions and behaviours are observable by other agents. Reasoning about information transparency is useful in many domains including security, privacy, and decision-making. In this paper, we present a formal framework for reasoning about information transparency properties in stochastic multi-agent systems. We extend Probabilistic Strategy Logic with new observability operators that capture the degree of observability of temporal properties by agents. We show that the model checking problem for the resulting logic is decidable.

A remarkable characteristic of quantum computing is the potential for reliable computation despite faulty qubits. This can be achieved through quantum error correction, which is typically implemented by repeatedly applying static syndrome checks, permitting correction of logical information. Recently, the development of time-dynamic approaches to error correction has uncovered new codes and new code implementations. In this work, we experimentally demonstrate three time-dynamic implementations of the surface code, each offering a unique solution to hardware design challenges and introducing flexibility in surface code realization. First, we embed the surface code on a hexagonal lattice, reducing the necessary couplings per qubit from four to three. Second, we walk a surface code, swapping the role of data and measure qubits each round, achieving error correction with built-in removal of accumulated non-computational errors. Finally, we realize the surface code using iSWAP gates instead of the traditional CNOT, extending the set of viable gates for error correction without additional overhead. We measure the error suppression factor when scaling from distance-3 to distance-5 codes of $\Lambda_{35,\text{hex}} = 2.15(2)$, $\Lambda_{35,\text{walk}} = 1.69(6)$, and $\Lambda_{35,\text{iSWAP}} = 1.56(2)$, achieving state-of-the-art error suppression for each. With detailed error budgeting, we explore their performance trade-offs and implications for hardware design. This work demonstrates that dynamic circuit approaches satisfy the demands for fault-tolerance and opens new alternative avenues for scalable hardware design., Comment: 11 pages, 5 figures, Supplementary Information

Quantum error correction is essential for bridging the gap between the error rates of physical devices and the extremely low logical error rates required for quantum algorithms. Recent error-correction demonstrations on superconducting processors have focused primarily on the surface code, which offers a high error threshold but poses limitations for logical operations. In contrast, the color code enables much more efficient logic, although it requires more complex stabilizer measurements and decoding techniques. Measuring these stabilizers in planar architectures such as superconducting qubits is challenging, and so far, realizations of color codes have not addressed performance scaling with code size on any platform. Here, we present a comprehensive demonstration of the color code on a superconducting processor, achieving logical error suppression and performing logical operations. Scaling the code distance from three to five suppresses logical errors by a factor of $\Lambda_{3/5}$ = 1.56(4). Simulations indicate this performance is below the threshold of the color code, and furthermore that the color code may be more efficient than the surface code with modest device improvements. Using logical randomized benchmarking, we find that transversal Clifford gates add an error of only 0.0027(3), which is substantially less than the error of an idling error correction cycle. We inject magic states, a key resource for universal computation, achieving fidelities exceeding 99% with post-selection (retaining about 75% of the data). Finally, we successfully teleport logical states between distance-three color codes using lattice surgery, with teleported state fidelities between 86.5(1)% and 90.7(1)%. This work establishes the color code as a compelling research direction to realize fault-tolerant quantum computation on superconducting processors in the near future.

Slums, informal settlements, and deprived areas are urban regions characterized by poverty. According to the United Nations, over one billion people reside in these areas, and this number is projected to increase. Additionally, these settlements are integral components of urban systems. We conducted a bibliometrical analysis and scoping study using the Web of Science Database to explore various perspectives on urban poverty, searching for scientific publications on the topic and providing details on the countries where the studies were conducted. Based on 3947 publications, we identify the extent to which domestic research organizations participate in studying urban poverty and which categories of science they investigate, including life sciences \& biomedicine, social sciences, technology, physical sciences, and arts & humanities. Thereby, we find that research on slums is often limited to specific countries, e.g. India, South Africa, Kenya, or Brazil. This focus is not necessarily correlated with the number of people living in slums. The scientific discourse is up to now predominantly shaped by medical and social sciences with few studies addressing technological questions. Finally, our analysis identifies several possible future directions for research on slums., Comment: 36 pages, 19 figures, 15 tables

Online reporting platforms have enabled citizens around the world to collectively share their opinions and report in real time on events impacting their local communities. Systematically organizing (e.g., categorizing by attributes) and geotagging large amounts of crowdsourced information is crucial to ensuring that accurate and meaningful insights can be drawn from this data and used by policy makers to bring about positive change. These tasks, however, typically require extensive manual annotation efforts. In this paper we present Uchaguzi-2022, a dataset of 14k categorized and geotagged citizen reports related to the 2022 Kenyan General Election containing mentions of election-related issues such as official misconduct, vote count irregularities, and acts of violence. We use this dataset to investigate whether language models can assist in scalably categorizing and geotagging reports, thus highlighting its potential application in the AI for Social Good space., Comment: COLING 2025

In this article, we present a long-duration autonomy approach for the control of connected and automated vehicles (CAVs) operating in a transportation network. In particular, we focus on the performance of CAVs at traffic bottlenecks, including roundabouts, merging roadways, and intersections. We take a principled approach based on optimal control, and derive a reactive controller with guarantees on safety, performance, and energy efficiency. We guarantee safety through high order control barrier functions (HOCBFs), which we ``lift'' to first order CBFs using time-optimal motion primitives. We demonstrate the performance of our approach in simulation and compare it to an optimal control-based approach., Comment: 7 pages, 4 figures

Scheduling virtual machines (VMs) to hosts in cloud data centers dictates efficiency and is an NP-hard problem with incomplete information. Prior work improved VM scheduling with predicted VM lifetimes. Our work further improves lifetime-aware scheduling using repredictions with lifetime distributions vs. one-shot prediction. The approach repredicts and adjusts VM and host lifetimes when incorrect predictions emerge. We also present novel approaches for defragmentation and regular system maintenance, which are essential to our data center reliability and optimizations, and are unexplored in prior work. We show that repredictions deliver a fundamental advance in effectiveness over one-shot prediction. We call our novel combination of distribution-based lifetime predictions and scheduling algorithms Lifetime Aware VM Allocation (LAVA). LAVA improves resource stranding and the number of empty hosts, which are critical for large VM scheduling, cloud system updates, and reducing dynamic energy consumption. Our approach runs in production within Google's hyperscale cloud data centers, where it improves efficiency by decreasing stranded compute and memory resources by ~3% and ~2% respectively, and increases availability for large VMs and cloud system updates by increasing empty hosts by 2.3-9.2 pp in production. We also show a reduction in VM migrations for host defragmentation and maintenance. In addition to our fleet-wide production deployment, we perform simulation studies to characterize the design space and show that our algorithm significantly outperforms the state of the art lifetime-based scheduling approach.

The Global Change Analysis Model (GCAM) simulates complex interactions between the coupled Earth and human systems, providing valuable insights into the co-evolution of land, water, and energy sectors under different future scenarios. Understanding the sensitivities and drivers of this multisectoral system can lead to more robust understanding of the different pathways to particular outcomes. The interactions and complexity of the coupled human-Earth systems make GCAM simulations costly to run at scale - a requirement for large ensemble experiments which explore uncertainty in model parameters and outputs. A differentiable emulator with similar predictive power, but greater efficiency, could provide novel scenario discovery and analysis of GCAM and its outputs, requiring fewer runs of GCAM. As a first use case, we train a neural network on an existing large ensemble that explores a range of GCAM inputs related to different relative contributions of energy production sources, with a focus on wind and solar. We complement this existing ensemble with interpolated input values and a wider selection of outputs, predicting 22,528 GCAM outputs across time, sectors, and regions. We report a median $R^2$ score of 0.998 for the emulator's predictions and an $R^2$ score of 0.812 for its input-output sensitivity., Comment: Presented at Tackling Climate Change with Machine Learning, NeurIPS 2024

We present our work on leveraging low-frame-rate monochrome (blue light) videos of fingertips, captured with an off-the-shelf fingerprint capture device, to extract vital signs and identify users. These videos utilize photoplethysmography (PPG), commonly used to measure vital signs like heart rate. While prior research predominantly utilizes high-frame-rate, multi-wavelength PPG sensors (e.g., infrared, red, or RGB), our preliminary findings demonstrate that both user identification and vital sign extraction are achievable with the low-frame-rate data we collected. Preliminary results are promising, with low error rates for both heart rate estimation and user authentication. These results indicate promise for effective biometric systems. We anticipate further optimization will enhance accuracy and advance healthcare and security., Comment: Accepted at the 2024 Asilomar Conference on Signals, Systems, and Computers. 5 pages, 5 figures, 2 tables

We give a canonical basis of theta functions for the Cox ring of two dimensional Looijenga pairs with affine interior, with structure constants naive counts of k-analytic disks in the total space of the universal deformation of the mirror (which, as this is dimension two, is isomorphic to the log Calabi-Yau surface itself)

The Ceresa cycle is a canonical algebraic $1$-cycle on the Jacobian of an algebraic curve. We construct an algorithm which, given a curve over a number field, often provides a certificate that the Ceresa cycle is non-torsion, without relying on the presence of any additional symmetries of the curve. Under the hypothesis that the Sato--Tate group is the whole of $\operatorname*{GSp}$, we prove that if the Ceresa class (the image of the Ceresa cycle in \'{e}tale cohomology) is non-torsion, then the algorithm will eventually terminate with a certificate attesting to this fact.

Challenge to control the quantum states of matter via light have been at the forefront of modern research on driven quantum matter. We explore the imprinting effects of structured light on superconductors, demonstrating how the quantum numbers of light-specifically spin angular momentum, orbital angular momentum, and radial order-can be transferred to the superconducting order parameter and control vortex dynamics. Using Laguerre-Gaussian beams, we show that by tuning the quantum numbers and the amplitude of the electric field, it is possible to manipulate a variety of vortex behaviors, including breathing vortex pairs, braiding vortex pairs, vortex droplets, supervortices, and swirling 2D vortex rings. More complex structure of vortex-clusters, such as vortex-flake structures, and standing wave motions, also emerge under specific quantum numbers. These results demonstrate the ability to control SC vortex motion and phase structures through structured light, offering potential applications in quantum fluids and optical control of superconducting states. Our findings present a diagram that links light's quantum numbers to the resulting SC vortex behaviors, highlighting the capacity of light to transfer its symmetry onto superconducting condensates. We point that this approach represents the extension of the printing to quantum printing by light in a coherent state of electrons.

Astronomers have found more than a dozen planets transiting 10-40 million year old stars, but even younger transiting planets have remained elusive. A possible reason for the lack of such discoveries is that newly formed planets are not yet in a configuration that would be recognized as a transiting planet or cannot exhibit transits because our view is blocked by a protoplanetary disk. However, we now know that many outer disks are warped; provided the inner disk is depleted, transiting planets may thus be visible. Here we report the observations of the transiting planet IRAS 04125+2902 b orbiting a 3 Myr, 0.7 M$_\odot$, pre-main sequence star in the Taurus Molecular Cloud. IRAS 04125+2902 hosts a nearly face-on (i $\sim$ 30$^\circ$) transitional disk and a wide binary companion. The planet has a period of 8.83 days, a radius of 10.9 R$_\oplus$ (0.97R$_J$), and a 95%-confidence upper limit on its mass of 90M$_\oplus$ (0.3M$_J$) from radial velocity measurements, making it a possible precursor of the super-Earths and sub-Neptunes that are commonly found around main-sequence stars. The rotational broadening of the star and the orbit of the wide (4", 635 AU) companion are both consistent with edge-on orientations. Thus, all components of the system appear to be aligned except the outer disk; the origin of this misalignment is unclear. Given the rare set of circumstances required to detect a transiting planet at ages when the disk is still present, IRAS 04125+2902 b likely provides a unique window into sub-Neptunes immediately following formation., Comment: Initial version submitted to Nature. Stellar, and hence planetary, parameters slightly differ from final version. Published version available at https://www.nature.com/articles/s41586-024-08123-3

Tropical cyclones (TCs) are highly dynamic natural disasters that travel vast distances and occupy a large spatial scale, leading to loss of life, economic strife, and destruction of infrastructure. The severe impact of TCs makes them crucial to monitor such that the collected data contributes to forecasting their trajectory and severity, as well as the provision of information to relief agencies. Among the various methods used to monitor TCs, Earth observation satellites are the most flexible, allowing for frequent observations with a wide variety of instruments. Traditionally, satellite scheduling algorithms assume nadir-directional observations, a limitation that can be alleviated by incorporating satellite agility and constellation reconfigurability -- two state-of-the-art concepts of operations (CONOPS) that extend the amount of time TCs can be observed from orbit. This paper conducts a systematic comparative analysis between both CONOPS to present the performance of each relative to baseline nadir-directional observations in monitoring TCs. A dataset of 100 historical TCs is used to provide a benchmark concerning real-world data through maximizing the number of quality observations. The results of the comparative analysis indicate that constellation reconfigurability allowing plane-change maneuvers outperforms satellite agility in the majority of TCs analyzed., Comment: 29 pages, 10 figures, Journal of Spacecraft and Rockets (accepted)

In the domain of large language models, considerable advancements have been attained in multimodal large language models and explainability research, propelled by the continuous technological progress and innovation. Nonetheless, security and privacy concerns continue to pose as prominent challenges in this field. The emergence of blockchain technology, marked by its decentralized nature, tamper-proof attributes, distributed storage functionality, and traceability, has provided novel approaches for resolving these issues. Both of these technologies independently hold vast potential for development; yet, their combination uncovers substantial cross-disciplinary opportunities and growth prospects. The current research tendencies are increasingly concentrating on the integration of blockchain with large language models, with the aim of compensating for their respective limitations through this fusion and promoting further technological evolution. In this study, we evaluate the advantages and developmental constraints of the two technologies, and explore the possibility and development potential of their combination. This paper primarily investigates the technical convergence in two directions: Firstly, the application of large language models to blockchain, where we identify six major development directions and explore solutions to the shortcomings of blockchain technology and their application scenarios; Secondly, the application of blockchain technology to large language models, leveraging the characteristics of blockchain to remedy the deficiencies of large language models and exploring its application potential in multiple fields.