Search

As our education systems continue to rebound from the COVID-19 pandemic we urge the next administration to prioritize investing wisely in the future of our nation through federal programs and policies that support the nation's 50 million K-12 public school students. A comprehensive and significant investment in the education of our public school students has never been more important to reducing the racial and socio-economic inequities plaguing American society today and providing a pathway for social and economic mobility. On behalf of public schools CEOs with deep expertise in education policy and as the largest employer in many communities, AASA crafts these recommendations for how the next administration can dramatically improve educational outcomes, reduce educational inequities and ensure the students in our schools today are able to excel in an increasingly competitive global workforce. Public schools perform critical functions for children and families, as well as for businesses, industries, and communities. Public schools are not just places where children can come to learn about democratic values; they are polling locations, interim food banks, and emergency shelters. The pandemic reminded us that the health and wellness of a community is directly linked to the health and wellness of the school system.

Foundation models have revolutionized artificial intelligence, setting new benchmarks in performance and enabling transformative capabilities across a wide range of vision and language tasks. However, despite the prevalence of spatio-temporal data in critical domains such as transportation, public health, and environmental monitoring, spatio-temporal foundation models (STFMs) have not yet achieved comparable success. In this paper, we articulate a vision for the future of STFMs, outlining their essential characteristics and the generalization capabilities necessary for broad applicability. We critically assess the current state of research, identifying gaps relative to these ideal traits, and highlight key challenges that impede their progress. Finally, we explore potential opportunities and directions to advance research towards the aim of effective and broadly applicable STFMs.

Continual Reinforcement Learning (CRL) is essential for developing agents that can learn, adapt, and accumulate knowledge over time. However, a fundamental challenge persists as agents must strike a delicate balance between plasticity, which enables rapid skill acquisition, and stability, which ensures long-term knowledge retention while preventing catastrophic forgetting. In this paper, we introduce SSDE, a novel structure-based approach that enhances plasticity through a fine-grained allocation strategy with Structured Sparsity and Dormant-guided Exploration. SSDE decomposes the parameter space into forward-transfer (frozen) parameters and task-specific (trainable) parameters. Crucially, these parameters are allocated by an efficient co-allocation scheme under sparse coding, ensuring sufficient trainable capacity for new tasks while promoting efficient forward transfer through frozen parameters. However, structure-based methods often suffer from rigidity due to the accumulation of non-trainable parameters, limiting exploration and adaptability. To address this, we further introduce a sensitivity-guided neuron reactivation mechanism that systematically identifies and resets dormant neurons, which exhibit minimal influence in the sparse policy network during inference. This approach effectively enhance exploration while preserving structural efficiency. Extensive experiments on the CW10-v1 Continual World benchmark demonstrate that SSDE achieves state-of-the-art performance, reaching a success rate of 95%, surpassing prior methods significantly in both plasticity and stability trade-offs (code is available at: https://github.com/chengqiArchy/SSDE).

Compared to traditional electromagnetic stealth (ES) materials, which are effective only within specific frequencies and orientations, intelligent reflecting surface (IRS) technology introduces a novel paradigm for achieving dynamic and adaptive ES by adapting its reflection pattern in real time to neutralize radar probing signals echoed back from the target. In this letter, we study an IRS-aided ES system mounted on an aerial target to evade radar detection admist uncertain/moving radar positions over an extended area. Specifically, we aim to optimize the IRS's passive reflection to minimize the maximum received signal-to-noise ratio (SNR) of the target echo signal in the area. A semi-closed-form solution is derived by first discretizing the continuous spatial frequency deviation to approximate the semi-infinite reflection gain constraint and then leveraging the Lagrange dual method. Simulation results are provided to validate that the proposed IRS-aided ES strategy can consistently reduce the reflection gains for radars located across a large region., Comment: 5 pages, 4 figures

Ongoing demands for smaller and more energy efficient electronic devices necessitate alternative interconnect materials with lower electrical resistivity at reduced dimensions. Despite the emergence of many promising candidates, synthesizing high quality nanostructures remains a major bottleneck in evaluating their performance. Here, we report the successful synthesis of Weyl semimetal NbAs nanowires via thermomechanical nanomolding, achieving single crystallinity and controlled diameters as small as 40 nm. Our NbAs nanowires exhibit a remarkably low room-temperature resistivity of 9.7 +/- 1.6 microOhm-cm, which is three to four times lower than their bulk counterpart. Theoretical calculations corroborate the experimental observations, attributing this exceptional resistivity reduction to surface dominant conduction with long carrier lifetime at finite temperatures. Further characterization of NbAs nanowires and bulk single crystals reveals high breakdown current density, robust stability, and superior thermal conductivity. Collectively, these properties highlight the strong potential of NbAs nanowires as next-generation interconnects, which can surpass the limitations of current copper-based interconnects. Technologically, our findings present a practical application of topological materials, while scientifically showcasing the fundamental properties uniquely accessible in nanoscale platforms., Comment: 5 figures

Zero-Knowledge Succinct Non-Interactive Argument of Knowledge (zk-SNARK) schemes have gained significant adoption in privacy-preserving applications, decentralized systems (e.g., blockchain), and verifiable computation due to their efficiency. However, the most efficient zk-SNARKs often rely on a one-time trusted setup to generate a public parameter, often known as the ``Powers of Tau" (PoT) string. The leakage of the secret parameter, $\tau$, in the string would allow attackers to generate false proofs, compromising the soundness of all zk-SNARK systems built on it. Prior proposals for decentralized setup ceremonies have utilized blockchain-based smart contracts to allow any party to contribute randomness to $\tau$ while also preventing censorship of contributions. For a PoT string of $d$-degree generated by the randomness of $m$ contributors, these solutions required a total of $O(md)$ on-chain operations (i.e., in terms of both storage and cryptographic operations). These operations primarily consisted of costly group operations, particularly scalar multiplication on pairing curves, which discouraged participation and limited the impact of decentralization In this work, we present Lite-PoT, which includes two key protocols designed to reduce participation costs: \emph{(i)} a fraud-proof protocol to reduce the number of expensive on-chain cryptographic group operations to $O(1)$ per contributor. Our experimental results show that (with one transaction per update) our protocol enables decentralized ceremonies for PoT strings up to a $2^{15}$ degree, an $\approx 16x$ improvement over existing on-chain solutions; \emph{(ii)} a proof aggregation technique that batches $m$ randomness contributions into one on-chain update with only $O(d)$ on-chain operations, independent of $m$. This significantly reduces the monetary cost of on-chain updates by $m$-fold via amortization.

Semi-supervised clustering is a basic problem in various applications. Most existing methods require knowledge of the ideal cluster number, which is often difficult to obtain in practice. Besides, satisfying the must-link constraints is another major challenge for these methods. In this work, we view the semi-supervised clustering task as a partitioning problem on a graph associated with the given dataset, where the similarity matrix includes a scaling parameter to reflect the must-link constraints. Utilizing a relaxation technique, we formulate the graph partitioning problem into a continuous optimization model that does not require the exact cluster number, but only an overestimate of it. We then propose a block coordinate descent algorithm to efficiently solve this model, and establish its convergence result. Based on the obtained solution, we can construct the clusters that theoretically meet the must-link constraints under mild assumptions. Furthermore, we verify the effectiveness and efficiency of our proposed method through comprehensive numerical experiments.

Gaussian Process (GP) models are widely utilized as surrogate models in scientific and engineering fields. However, standard GP models are limited to continuous variables due to the difficulties in establishing correlation structures for categorical variables. To overcome this limitati on, we introduce WEighted Euclidean distance matrices Gaussian Process (WEGP). WEGP constructs the kernel function for each categorical input by estimating the Euclidean distance matrix (EDM) among all categorical choices of this input. The EDM is represented as a linear combination of several predefined base EDMs, each scaled by a positive weight. The weights, along with other kernel hyperparameters, are inferred using a fully Bayesian framework. We analyze the predictive performance of WEGP theoretically. Numerical experiments validate the accuracy of our GP model, and by WEGP, into Bayesian Optimization (BO), we achieve superior performance on both synthetic and real-world optimization problems.

Pulsar wind nebulae (PWNe) are important sources for understanding galactic high-energy processes, but it is controversial until now about how high-energy particles in PWNe are accelerated and transported. Lacking radio counterparts of X-ray PWNe (the proposed acceleration sites) introduce difficulties to better understandings in multi wavelengths. Our recent 3, 6, and 16\,cm high-resolution observations of G11.2$-$0.3 PWN with the Australia Telescope Compact Array (ATCA) uniquely show morphological similarity with its X-ray PWN (a torus/jet feature). Spectral indices of the radio torus and jet are around -0.09 and -0.10, respectively. Meanwhile for the jet region, the spectral break between radio and X-ray spectra implies particle acceleration mechanisms other than a diffusive shock acceleration. Polarization results suggest a helical B-field inside the jet, the equipartition B-field strength of which is below 100\,$\mu$G., Comment: 15 pages, 13 figures

Type-I X-ray bursts are thermonuclear explosions caused by the unstable burning of accreted material on the surface of neutron stars. We report the detection of seven type-I X-ray bursts from the ultracompact X-ray binary M15 X-2 observed by the Neutron Star Interior Composition Explorer (NICER) during its 2022 outburst. We found all the bursts occurred in the soft state and exhibited similar light curve profiles, with no cases of photospheric radius expansion. Time-resolved spectroscopy showed clear deviations from the blackbody model during the first ten seconds of all the bursts. The fits were improved by using the enhanced persistent emission method, which we interpret as evidence of burst-disk interaction. We compared the performance of these models against a neutron star atmosphere model and found it made no significant improvements. After analyzing the burst rise times and fuel composition, we propose that these bursts were powered by the burning of pure helium, confirming the ultracompact nature of the source.

Regularizing neural networks is important for anticipating model behavior in regions of the data space that are not well represented. In this work, we propose a regularization technique for enforcing a level of smoothness in the mapping between the data input space and the loss value. We specify the level of regularity by requiring that the loss of the network satisfies an elliptic operator over the data domain. To do this, we modify the usual empirical risk minimization objective such that we instead minimize a new objective that satisfies an elliptic operator over points within the domain. This allows us to use existing theory on elliptic operators to anticipate the behavior of the error for points outside the training set. We propose a tractable computational method that approximates the behavior of the elliptic operator while being computationally efficient. Finally, we analyze the properties of the proposed regularization to understand the performance on common problems of distribution shift and group imbalance. Numerical experiments confirm the utility of the proposed regularization technique., Comment: ICLR 2025

Large Language Models (LLMs) have reshaped code generation by synergizing their exceptional comprehension of natural language and programming syntax, thereby substantially boosting developer productivity. These advancements have prompted numerous efforts to quantitatively evaluate their coding capabilities. However, persistent challenges, such as benchmark leakage, data dissipation, and limited system accessibility, continue to impede a timely and accurate assessment. To address these limitations, we introduce CodeArena, an online evaluation framework tailored for LLM code generation. The key innovation is a collective evaluation mechanism, which dynamically recalibrates individual model scores based on the holistic performance of all participating models, mitigating score biases caused by widespread benchmark leakage. In addition, CodeArena ensures open access to all submitted solutions and test cases and provides automation-friendly APIs to streamline the code evaluation workflow. Our main contributions are: (1) a collective evaluation system for unbiased assessment, (2) a public repository of solutions and test cases, and (3) automation-ready APIs for seamless integration.

Structure reasoning is a fundamental capability of large language models (LLMs), enabling them to reason about structured commonsense and answer multi-hop questions. However, existing benchmarks for structure reasoning mainly focus on horizontal and coordinate structures (\emph{e.g.} graphs), overlooking the hierarchical relationships within them. Hierarchical structure reasoning is crucial for human cognition, particularly in memory organization and problem-solving. It also plays a key role in various real-world tasks, such as information extraction and decision-making. To address this gap, we propose HiBench, the first framework spanning from initial structure generation to final proficiency assessment, designed to benchmark the hierarchical reasoning capabilities of LLMs systematically. HiBench encompasses six representative scenarios, covering both fundamental and practical aspects, and consists of 30 tasks with varying hierarchical complexity, totaling 39,519 queries. To evaluate LLMs comprehensively, we develop five capability dimensions that depict different facets of hierarchical structure understanding. Through extensive evaluation of 20 LLMs from 10 model families, we reveal key insights into their capabilities and limitations: 1) existing LLMs show proficiency in basic hierarchical reasoning tasks; 2) they still struggle with more complex structures and implicit hierarchical representations, especially in structural modification and textual reasoning. Based on these findings, we create a small yet well-designed instruction dataset, which enhances LLMs' performance on HiBench by an average of 88.84\% (Llama-3.1-8B) and 31.38\% (Qwen2.5-7B) across all tasks. The HiBench dataset and toolkit are available here, https://github.com/jzzzzh/HiBench, to encourage evaluation.

In order to find closed form solutions of nonintegrable nonlinear ordinary differential equations, numerous tricks have been proposed. The goal of this short review is to recall classical, 19th-century results, completed in 2006 by Eremenko, which can be turned into algorithms, thus avoiding \textit{ad hoc} assumptions, able to provide \textit{all} (as opposed to some) solutions in a precise class. To illustrate these methods, we present some new such exact solutions, physically relevent., Comment: 11 pages, 1 figure, to appear, Advances in mathematics (China)

Disentangled representation learning aims to uncover latent variables underlying the observed data, and generally speaking, rather strong assumptions are needed to ensure identifiability. Some approaches rely on sufficient changes on the distribution of latent variables indicated by auxiliary variables such as domain indices, but acquiring enough domains is often challenging. Alternative approaches exploit structural sparsity assumptions on the mixing procedure, but such constraints are usually (partially) violated in practice. Interestingly, we find that these two seemingly unrelated assumptions can actually complement each other to achieve identifiability. Specifically, when conditioned on auxiliary variables, the sparse mixing procedure assumption provides structural constraints on the mapping from estimated to true latent variables and hence compensates for potentially insufficient distribution changes. Building on this insight, we propose an identifiability theory with less restrictive constraints regarding distribution changes and the sparse mixing procedure, enhancing applicability to real-world scenarios. Additionally, we develop an estimation framework incorporating a domain encoding network and a sparse mixing constraint and provide two implementations based on variational autoencoders and generative adversarial networks, respectively. Experiment results on synthetic and real-world datasets support our theoretical results.

In this paper, we conduct a detailed study on the effect of Radiative Torque Disruption (RATD) mechanism on the fragmentation of micrometer-sized dust grains into nanoparticles within the heliosphere. We start by estimating the disruption timescales for dust grains under various centrifugal stresses. Our numerical calculations demonstrate that RATD is a highly effective mechanism for breaking down micrometer-sized grains, producing nanoparticles more efficiently than other fragmentation processes. RATD also prevents micrometer-sized grains from being expelled by radiation pressure. Our findings indicate that the location of the present water snow line depends not only on temperature but also on the size of dust grains. For smaller grains, the snow line can shift outward beyond the position defined by thermal sublimation. Furthermore, we model the size distribution of dust grains modified by the RATD mechanism using a simplified model, showing that rotational disruption significantly decreases the number density of micrometer-sized grains while substantially increasing the number density of sub-micrometer-sized grains. However, the fraction of dust grains aligned at high-$J$ attractors by radiative torques less than 80\% can considerably weaken the effect of RATD on the grain size distribution. Finally, we suggest several experiments that could potentially test the RATD mechanism and discuss the uncertainties of our model in more realistic applications to heliospheric dust.

We present the timing and spectral analyses of the NICER, NuSTAR, and IXPE observations of the magnetar 1E 1841-045 covering 82 days following its August 2024 bursting activity as well as radio observations utilizing MeerKAT and Effelsberg. We supplement our study with a historical NuSTAR and all 2024 pre-outburst NICER observations. The outburst is marked by an X-ray flux enhancement of a factor 1.6 compared to the historical level, predominantly driven by a newly-formed non-thermal emitting component with a photon index $\Gamma=1.5$. This flux showed a 20% decay at the end of our monitoring campaign. The radio monitoring did not reveal any pulsed radio emission with an upper-limit of 20 mJy and 50 mJy ms on the mean flux density and single pulse fluence, respectively. We detect a spin-up glitch at outburst onset with a $\Delta\nu=6.1\times10^{-8}$ Hz and a $\Delta\dot{\nu}=-1.4\times10^{-14}$ Hz s$^{-1}$, consistent with the near-universality of this behavior among the continuously-monitored magnetars. Most intriguingly, the 1E 1841-045 2-10 keV pulse profile is markedly different compared to pre-outburst; it shows a new, narrow (0.1 cycles) peak that appears to shift towards merging with the main, persistently-present, pulse. This is the second case of pulse-peak migration observed in magnetars after SGR 1830$-$0645, and the two sources exhibit a similar rate of phase shift. This implies that this phenomenon is not unique and might present itself in the broader population. The newly-formed peak for 1E 1841-045 is non-thermal, with emission extending to $\gtrsim20$ keV, in contrast to the case of SGR 1830$-$0645. Our results are consistent with an untwisting magnetic field bundle with migration towards the magnetic pole, perhaps accompanied by plastic motion of the crust., Comment: 16 pages, 6 figures, 4 tables, submitted to ApJ, comments welcome

Anisotropy plays a critical role in governing the mechanical, thermal, electrical, magnetic, and optical properties of materials, influencing their behavior across diverse applications. Probing and quantifying this directional dependence is crucial for advancing materials science and biomedical research, as it provides a deeper understanding of structural orientations at the molecular level, encompassing both scientific and industrial benefits. This study introduces the "4+ Angle Polarization" widget, an innovative extension to the open-source Quasar platform (https://quasar.codes/), tailored for advanced multiple-angle polarization analysis. This toolbox enables precise molecular orientation analysis of complex microspectroscopic datasets through a streamlined workflow. Using polarized Fourier transform infrared (p-FTIR) spectroscopy, we demonstrate its versatility across various sample types, including polylactic acid (PLA) organic crystals, murine cortical bone, and human osteons. By overcoming the limitations of traditional two-angle methods, the widget significantly enhances the accuracy of structural and orientational analysis. This novel analytical tool expands the potential of multiple-angle p-FTIR techniques into advanced characterization of structural anisotropy in heterogeneous systems, providing transformative insights for materials characterization, biomedical imaging and beyond., Comment: 20 pages, 9 figures

Large Language Models (LLMs) have significantly advanced the fact-checking studies. However, existing automated fact-checking evaluation methods rely on static datasets and classification metrics, which fail to automatically evaluate the justification production and uncover the nuanced limitations of LLMs in fact-checking. In this work, we introduce FACT-AUDIT, an agent-driven framework that adaptively and dynamically assesses LLMs' fact-checking capabilities. Leveraging importance sampling principles and multi-agent collaboration, FACT-AUDIT generates adaptive and scalable datasets, performs iterative model-centric evaluations, and updates assessments based on model-specific responses. By incorporating justification production alongside verdict prediction, this framework provides a comprehensive and evolving audit of LLMs' factual reasoning capabilities, to investigate their trustworthiness. Extensive experiments demonstrate that FACT-AUDIT effectively differentiates among state-of-the-art LLMs, providing valuable insights into model strengths and limitations in model-centric fact-checking analysis.

We present the results of a search for nonlinear gravitational wave memory in the NANOGrav 15-year data set. We find no significant evidence for memory signals in the dataset, with a maximum Bayes factor of 3.1 in favor of a model including memory. We therefore place upper limits on the strain of potential gravitational wave memory events as a function of sky location and observing epoch. We find upper limits that are not always more constraining than previous NANOGrav results. We show that it is likely due to the increase in common red noise between the 12.5-year and 15-year NANOGrav datasets.

The Kerr separatrix is a boundary in parameter space that separates bound orbits from plunging orbits in the Kerr black hole space-time. Recently, Stein and Warburton found a polynomial equation for the location of the separatrix, for two different choices of inclination parameter. Following a method of Levin and Perez-Giz developed for the equatorial case, we use a correspondence between homoclinic orbits and unstable spherical orbits to derive explicit solutions to the separatrix polynomials. These solutions are parametrised in terms of the radius of the unstable spherical orbit., Comment: 12 pages, accepted for publication in Classical and Quantum Gravity

This paper investigates integrated localization and communication in a multi-cell system and proposes a coordinated beamforming algorithm to enhance target localization accuracy while preserving communication performance. Within this integrated sensing and communication (ISAC) system, the Cramer-Rao lower bound (CRLB) is adopted to quantify the accuracy of target localization, with its closed-form expression derived for the first time. It is shown that the nuisance parameters can be disregarded without impacting the CRLB of time of arrival (TOA)-based target localization. Capitalizing on the derived CRLB, we formulate a nonconvex coordinated beamforming problem to minimize the CRLB while satisfying signal-to-interference-plus-noise ratio (SINR) constraints in communication. To facilitate the development of solution, we reformulate the original problem into a more tractable form and solve it through semi-definite programming (SDP). Notably, we show that the proposed algorithm can always obtain rank-one global optimal solutions under mild conditions. Finally, numerical results demonstrate the superiority of the proposed algorithm over benchmark algorithms and reveal the performance trade-off between localization accuracy and communication SINR.

Iterative data generation and model retraining are widely used to align large language models (LLMs). It typically involves a policy model to generate on-policy responses and a reward model to guide training data selection. Direct Preference Optimization (DPO) further enhances this process by constructing preference pairs of chosen and rejected responses. In this work, we aim to \emph{scale up} the number of on-policy samples via repeated random sampling to improve alignment performance. Conventional practice selects the sample with the highest reward as chosen and the lowest as rejected for DPO. However, our experiments reveal that this strategy leads to a \emph{decline} in performance as the sample size increases. To address this, we investigate preference data construction through the lens of underlying normal distribution of sample rewards. We categorize the reward space into seven representative points and systematically explore all 21 ($C_7^2$) pairwise combinations. Through evaluations on four models using AlpacaEval 2, we find that selecting the rejected response at reward position $\mu - 2\sigma$ rather than the minimum reward, is crucial for optimal performance. We finally introduce a scalable preference data construction strategy that consistently enhances model performance as the sample scale increases.

Data valuation is increasingly used in machine learning (ML) to decide the fair compensation for data owners and identify valuable or harmful data for improving ML models. Cooperative game theory-based data valuation, such as Data Shapley, requires evaluating the data utility (e.g., validation accuracy) and retraining the ML model for multiple data subsets. While most existing works on efficient estimation of the Shapley values have focused on reducing the number of subsets to evaluate, our framework, \texttt{DUPRE}, takes an alternative yet complementary approach that reduces the cost per subset evaluation by predicting data utilities instead of evaluating them by model retraining. Specifically, given the evaluated data utilities of some data subsets, \texttt{DUPRE} fits a \emph{Gaussian process} (GP) regression model to predict the utility of every other data subset. Our key contribution lies in the design of our GP kernel based on the sliced Wasserstein distance between empirical data distributions. In particular, we show that the kernel is valid and positive semi-definite, encodes prior knowledge of similarities between different data subsets, and can be efficiently computed. We empirically verify that \texttt{DUPRE} introduces low prediction error and speeds up data valuation for various ML models, datasets, and utility functions., Comment: 16 pages, 7 figures, the paper got accepted AAMAS 2025

In this study, we present a comprehensive analysis of an unidentified point-like ultra-high-energy (UHE) $\gamma$-ray source, designated as 1LHAASO J1740+0948u, situated in the vicinity of the middle-aged pulsar PSR J1740+1000. The detection significance reached 17.1$\sigma$ (9.4$\sigma$) above 25$\,$TeV (100$\,$TeV). The source energy spectrum extended up to 300$\,$TeV, which was well fitted by a log-parabola function with $N0 = (1.93\pm0.23) \times 10^{-16} \rm{TeV^{-1}\,cm^{-2}\,s^{-2}}$, $\alpha = 2.14\pm0.27$, and $\beta = 1.20\pm0.41$ at E0 = 30$\,$TeV. The associated pulsar, PSR J1740+1000, resides at a high galactic latitude and powers a bow-shock pulsar wind nebula (BSPWN) with an extended X-ray tail. The best-fit position of the gamma-ray source appeared to be shifted by $0.2^{\circ}$ with respect to the pulsar position. As the (i) currently identified pulsar halos do not demonstrate such offsets, and (ii) centroid of the gamma-ray emission is approximately located at the extension of the X-ray tail, we speculate that the UHE $\gamma$-ray emission may originate from re-accelerated electron/positron pairs that are advected away in the bow-shock tail., Comment: Corrected spelling errors in several author names

Numerical simulations are essential to understand the complex physics accompanying the merger of binary systems of neutron stars. However, these simulations become computationally challenging when they have to model the merger remnants on timescales over which secular phenomena, such as the launching of magnetically driven outflows, develop. To tackle these challenges, we have recently developed a hybrid approach that combines, via a hand-off transition, a fully general-relativistic code (FIL) with a more efficient code making use of the conformally flat approximation (BHAC+). We here report important additional developments of BHAC+ consisting of the inclusion of gravitational-wave radiation-reaction contributions and of higher-order formulations of the equations of general-relativistic magnetohydrodynamics. Both improvements have allowed us to explore BNS merger remnants with high accuracy and over timescales that would have been computationally prohibitive otherwise. More specifically, we have investigated the impact of the magnetic-field strength on the long-term (\ie $\sim 200\,{\rm ms}$) and high-resolution (\ie $150\,{\rm m}$) evolutions of the ``magnetar'' resulting from the merger of two neutron stars with a realistic equation of state. In this way, and for sufficiently large magnetic fields, we observe the weakening or suppression of differential rotation and the generation of magnetic flares in the outer layers of the remnant. These flares, driven mostly by the Parker instability, are responsible for intense and collimated Poynting flux outbursts and mass ejections. This novel phenomenology offers the possibility of seeking corresponding signatures from the observations of short gamma-ray bursts and hence revealing the existence of a long-lived strongly magnetized remnant., Comment: 30 pages, 17 figures

Instruction Fine-tuning (IFT) can enhance the helpfulness of Large Language Models (LLMs), but it may lower their truthfulness. This trade-off arises because IFT steers LLMs to generate responses with long-tail knowledge that is not well covered during pre-training, leading to more informative but less truthful answers when generalizing to unseen tasks. In this paper, we empirically demonstrate this helpfulness-truthfulness trade-off in IFT and propose $\textbf{UNIT}$, a novel IFT paradigm to address it. UNIT teaches LLMs to recognize their uncertainty and explicitly reflect it at the end of their responses. Experimental results show that UNIT-tuned models maintain their helpfulness while distinguishing between certain and uncertain claims, thereby reducing hallucinations.

The treatment effect represents the average causal impact or outcome difference between treatment and control groups. Treatment effects can be estimated through social experiments, regression models, matching estimators, and instrumental variables. In this paper, we introduce a novel kernel-matching estimator for treatment effect estimation. This method is particularly beneficial in observational studies where randomized control trials are not feasible, as it uses the full sample to increase the efficiency and robustness of treatment effect estimates. We demonstrate that the proposed estimator is consistent and asymptotically efficient under certain conditions. Through Monte Carlo simulations, we show that the estimator performs favorably against other estimators in the literature. Finally, we apply our method to data from the National Supported Work Demonstration to illustrate its practical application.

Deepfakes, particularly those involving faceswap-based manipulations, have sparked significant societal concern due to their increasing realism and potential for misuse. Despite rapid advancements in generative models, detection methods have not kept pace, creating a critical gap in defense strategies. This disparity is further amplified by the disconnect between academic research and real-world applications, which often prioritize different objectives and evaluation criteria. In this study, we take a pivotal step toward bridging this gap by presenting a novel observation: the post-processing step of super-resolution, commonly employed in real-world scenarios, substantially undermines the effectiveness of existing deepfake detection methods. To substantiate this claim, we introduce and publish the first real-world faceswap dataset, collected from popular online faceswap platforms. We then qualitatively evaluate the performance of state-of-the-art deepfake detectors on real-world deepfakes, revealing that their accuracy approaches the level of random guessing. Furthermore, we quantitatively demonstrate the significant performance degradation caused by common post-processing techniques. By addressing this overlooked challenge, our study underscores a critical avenue for enhancing the robustness and practical applicability of deepfake detection methods in real-world settings.

Leveraging Large Language Models (LLMs) for generative recommendation has attracted significant research interest, where item tokenization is a critical step. It involves assigning item identifiers for LLMs to encode user history and generate the next item. Existing approaches leverage either token-sequence identifiers, representing items as discrete token sequences, or single-token identifiers, using ID or semantic embeddings. Token-sequence identifiers face issues such as the local optima problem in beam search and low generation efficiency due to step-by-step generation. In contrast, single-token identifiers fail to capture rich semantics or encode Collaborative Filtering (CF) information, resulting in suboptimal performance. To address these issues, we propose two fundamental principles for item identifier design: 1) integrating both CF and semantic information to fully capture multi-dimensional item information, and 2) designing order-agnostic identifiers without token dependency, mitigating the local optima issue and achieving simultaneous generation for generation efficiency. Accordingly, we introduce a novel set identifier paradigm for LLM-based generative recommendation, representing each item as a set of order-agnostic tokens. To implement this paradigm, we propose SETRec, which leverages CF and semantic tokenizers to obtain order-agnostic multi-dimensional tokens. To eliminate token dependency, SETRec uses a sparse attention mask for user history encoding and a query-guided generation mechanism for simultaneous token generation. We instantiate SETRec on T5 and Qwen (from 1.5B to 7B). Extensive experiments demonstrate its effectiveness under various scenarios (e.g., full ranking, warm- and cold-start ranking, and various item popularity groups). Moreover, results validate SETRec's superior efficiency and show promising scalability on cold-start items as model sizes increase.

Data augmentation is an important technique in training deep neural networks as it enhances their ability to generalize and remain robust. While data augmentation is commonly used to expand the sample size and act as a consistency regularization term, there is a lack of research on the relationship between them. To address this gap, this paper introduces a more comprehensive mathematical framework for data augmentation. Through this framework, we establish that the expected risk of the shifted population is the sum of the original population risk and a gap term, which can be interpreted as a consistency regularization term. The paper also provides a theoretical understanding of this gap, highlighting its negative effects on the early stages of training. We also propose a method to mitigate these effects. To validate our approach, we conducted experiments using same data augmentation techniques and computing resources under several scenarios, including standard training, out-of-distribution, and imbalanced classification. The results demonstrate that our methods surpass compared methods under all scenarios in terms of generalization ability and convergence stability. We provide our code implementation at the following link: https://github.com/ydlsfhll/ASPR.

Catheter ablation of Atrial Fibrillation (AF) consists of a one-size-fits-all treatment with limited success in persistent AF. This may be due to our inability to map the dynamics of AF with the limited resolution and coverage provided by sequential contact mapping catheters, preventing effective patient phenotyping for personalised, targeted ablation. Here we introduce FibMap, a graph recurrent neural network model that reconstructs global AF dynamics from sparse measurements. Trained and validated on 51 non-contact whole atria recordings, FibMap reconstructs whole atria dynamics from 10% surface coverage, achieving a 210% lower mean absolute error and an order of magnitude higher performance in tracking phase singularities compared to baseline methods. Clinical utility of FibMap is demonstrated on real-world contact mapping recordings, achieving reconstruction fidelity comparable to non-contact mapping. FibMap's state-spaces and patient-specific parameters offer insights for electrophenotyping AF. Integrating FibMap into clinical practice could enable personalised AF care and improve outcomes., Comment: Under review

Online continual learning for image classification is crucial for models to adapt to new data while retaining knowledge of previously learned tasks. This capability is essential to address real-world challenges involving dynamic environments and evolving data distributions. Traditional approaches predominantly employ Convolutional Neural Networks, which are limited to processing images as grids and primarily capture local patterns rather than relational information. Although the emergence of transformer architectures has improved the ability to capture relationships, these models often require significantly larger resources. In this paper, we present a novel online continual learning framework based on Graph Attention Networks (GATs), which effectively capture contextual relationships and dynamically update the task-specific representation via learned attention weights. Our approach utilizes a pre-trained feature extractor to convert images into graphs using hierarchical feature maps, representing information at varying levels of granularity. These graphs are then processed by a GAT and incorporate an enhanced global pooling strategy to improve classification performance for continual learning. In addition, we propose the rehearsal memory duplication technique that improves the representation of the previous tasks while maintaining the memory budget. Comprehensive evaluations on benchmark datasets, including SVHN, CIFAR10, CIFAR100, and MiniImageNet, demonstrate the superiority of our method compared to the state-of-the-art methods., Comment: 16 pages

We introduce a density model for random quotients of a free product of finitely generated groups. We prove that a random quotient in this model has the following properties with overwhelming probability: if the density is below $1/2$, the free factors embed into the random quotient and the random quotient is hyperbolic relative to the free factors. Further, there is a phase transition at $1/2$, with the random quotient being a finite group above this density. If the density is below $1/6$, the random quotient is cubulated relative to the free factors. Moreover, if the free factors are cubulated, then so is the random quotient., Comment: 43 pages, 9 figures

A language can have different varieties. These varieties can affect the performance of natural language processing (NLP) models, including large language models (LLMs), which are often trained on data from widely spoken varieties. This paper introduces a novel and cost-effective approach to benchmark model performance across language varieties. We argue that international online review platforms, such as Booking.com, can serve as effective data sources for constructing datasets that capture comments in different language varieties from similar real-world scenarios, like reviews for the same hotel with the same rating using the same language (e.g., Mandarin Chinese) but different language varieties (e.g., Taiwan Mandarin, Mainland Mandarin). To prove this concept, we constructed a contextually aligned dataset comprising reviews in Taiwan Mandarin and Mainland Mandarin and tested six LLMs in a sentiment analysis task. Our results show that LLMs consistently underperform in Taiwan Mandarin., Comment: Accepted by 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), theme track

Kolmogorov--Arnold networks (KANs) have demonstrated their potential as an alternative to multi-layer perceptions (MLPs) in various domains, especially for science-related tasks. However, transfer learning of KANs remains a relatively unexplored area. In this paper, inspired by Tucker decomposition of tensors and evidence on the low tensor-rank structure in KAN parameter updates, we develop low tensor-rank adaptation (LoTRA) for fine-tuning KANs. We study the expressiveness of LoTRA based on Tucker decomposition approximations. Furthermore, we provide a theoretical analysis to select the learning rates for each LoTRA component to enable efficient training. Our analysis also shows that using identical learning rates across all components leads to inefficient training, highlighting the need for an adaptive learning rate strategy. Beyond theoretical insights, we explore the application of LoTRA for efficiently solving various partial differential equations (PDEs) by fine-tuning KANs. Additionally, we propose Slim KANs that incorporate the inherent low-tensor-rank properties of KAN parameter tensors to reduce model size while maintaining superior performance. Experimental results validate the efficacy of the proposed learning rate selection strategy and demonstrate the effectiveness of LoTRA for transfer learning of KANs in solving PDEs. Further evaluations on Slim KANs for function representation and image classification tasks highlight the expressiveness of LoTRA and the potential for parameter reduction through low tensor-rank decomposition.

The core-collapse supernova remnant (SNR) Cassiopeia A (Cas A) is one of the brightest galactic radio sources with an angular radius of $\sim$ 2.5 $\arcmin$. Although no extension of this source has been detected in the $\gamma$-ray band, using more than 1000 days of LHAASO data above $\sim 0.8$ TeV, we find that its spectrum is significantly softer than those obtained with Imaging Air Cherenkov Telescopes (IACTs) and its flux near $\sim 1$ TeV is about two times higher. In combination with analyses of more than 16 years of \textit{Fermi}-LAT data covering $0.1 \, \mathrm{GeV} - 1 \, \mathrm{TeV}$, we find that the spectrum above 30 GeV deviates significantly from a single power-law, and is best described by a smoothly broken power-law with a spectral index of $1.90 \pm 0.15_\mathrm{stat}$ ($3.41 \pm 0.19_\mathrm{stat}$) below (above) a break energy of $0.63 \pm 0.21_\mathrm{stat} \, \mathrm{TeV}$. Given differences in the angular resolution of LHAASO-WCDA and IACTs, TeV $\gamma$-ray emission detected with LHAASO may have a significant contribution from regions surrounding the SNR illuminated by particles accelerated earlier, which, however, are treated as background by IACTs. Detailed modelling can be used to constrain acceleration processes of TeV particles in the early stage of SNR evolution.

A fundamental issue in deep learning has been adversarial robustness. As these systems have scaled, such issues have persisted. Currently, large language models (LLMs) with billions of parameters suffer from adversarial attacks just like their earlier, smaller counterparts. However, the threat models have changed. Previously, having gray-box access, where input embeddings or output logits/probabilities were visible to the user, might have been reasonable. However, with the introduction of closed-source models, no information about the model is available apart from the generated output. This means that current black-box attacks can only utilize the final prediction to detect if an attack is successful. In this work, we investigate and demonstrate the potential of attack guidance, akin to using output probabilities, while having only black-box access in a classification setting. This is achieved through the ability to elicit confidence from the model. We empirically show that the elicited confidence is calibrated and not hallucinated for current LLMs. By minimizing the elicited confidence, we can therefore increase the likelihood of misclassification. Our new proposed paradigm demonstrates promising state-of-the-art results on three datasets across two models (LLaMA-3-8B-Instruct and Mistral-7B-Instruct-V0.3) when comparing our technique to existing hard-label black-box attack methods that introduce word-level substitutions., Comment: Published in ICLR 2025. The code is publicly available at https://github.com/Aniloid2/Confidence_Elicitation_Attacks

Phishing attacks remain a significant threat in the digital age, yet organizations lack effective methods to tackle phishing attacks without leaking sensitive information. Phish bowl initiatives are a vital part of cybersecurity efforts against these attacks. However, traditional phish bowls require manual anonymization and are often limited to internal use. To overcome these limitations, we introduce AdaPhish, an AI-powered phish bowl platform that automatically anonymizes and analyzes phishing emails using large language models (LLMs) and vector databases. AdaPhish achieves real-time detection and adaptation to new phishing tactics while enabling long-term tracking of phishing trends. Through automated reporting, adaptive analysis, and real-time alerts, AdaPhish presents a scalable, collaborative solution for phishing detection and cybersecurity education., Comment: 7 pages, 3 figures, 2 tables, accepted in 4th IEEE International Conference on AI in Cybersecurity (ICAIC)

Quantum key distribution requires tight and reliable bounds on the secret key rate to ensure robust security. This is particularly so for the regime of finite block sizes, where the optimization of generalized Renyi entropic quantities is known to provide tighter bounds on the key rate. However, such an optimization is often non-trivial, and the non-monotonicity of the key rate in terms of the Renyi parameter demands additional optimization to determine the optimal Renyi parameter as a function of block sizes. In this work, we present a tight analytical bound on the Renyi entropy in terms of the Renyi divergence and derive the analytical gradient of the Renyi divergence. This enables us to generalize existing state-of-the-art numerical frameworks for the optimization of the key rate. With this generalized framework, we show improvements in regimes of high loss and low block sizes, which are particularly relevant for long-distance satellite-based protocols., Comment: 6+5, 4 figures, comments are welcomed

Social media is a primary medium for information diffusion during natural disasters. The social media ecosystem has been used to identify destruction, analyze opinions and organize aid. While the overall picture and aggregate trends may be important, a crucial part of the picture is the connections on these sites. These bridges are essential to facilitate information flow within the network. In this work, we perform a multi-platform analysis (X, Reddit, YouTube) of Hurricanes Helene and Milton, which occurred in quick session to each other in the US in late 2024. We construct network graphs to understand the properties of effective bridging content and users. We find that bridges tend to exist on X, that bridging content is complex, and that bridging users have relatable affiliations related to gender, race and job. Public organizations can use these characteristics to manage their social media personas during natural disasters more effectively.

Learning-to-Defer (L2D) facilitates optimal task allocation between AI systems and decision-makers. Despite its potential, we show that current two-stage L2D frameworks are highly vulnerable to adversarial attacks, which can misdirect queries or overwhelm decision agents, significantly degrading system performance. This paper conducts the first comprehensive analysis of adversarial robustness in two-stage L2D frameworks. We introduce two novel attack strategies -- untargeted and targeted -- that exploit inherent structural vulnerabilities in these systems. To mitigate these threats, we propose SARD, a robust, convex, deferral algorithm rooted in Bayes and $(\mathcal{R},\mathcal{G})$-consistency. Our approach guarantees optimal task allocation under adversarial perturbations for all surrogates in the cross-entropy family. Extensive experiments on classification, regression, and multi-task benchmarks validate the robustness of SARD.

Existing methods for analyzing linguistic content from picture descriptions for assessment of cognitive-linguistic impairment often overlook the participant's visual narrative path, which typically requires eye tracking to assess. Spatio-semantic graphs are a useful tool for analyzing this narrative path from transcripts alone, however they are limited by the need for manual tagging of content information units (CIUs). In this paper, we propose an automated approach for estimation of spatio-semantic graphs (via automated extraction of CIUs) from the Cookie Theft picture commonly used in cognitive-linguistic analyses. The method enables the automatic characterization of the visual semantic path during picture description. Experiments demonstrate that the automatic spatio-semantic graphs effectively differentiate between cognitively impaired and unimpaired speakers. Statistical analyses reveal that the features derived by the automated method produce comparable results to the manual method, with even greater group differences between clinical groups of interest. These results highlight the potential of the automated approach for extracting spatio-semantic features in developing clinical speech models for cognitive impairment assessment., Comment: To appear in ICASSP 2025

Large Language Models (LLMs) have the potential to revolutionize data analytics by simplifying tasks such as data discovery and SQL query synthesis through natural language interactions. This work serves as a pivotal first step toward the development of foundation models explicitly designed for data analytics applications. To propel this vision forward, we unveil a new data recipe for post-training LLMs, enhancing their comprehension of data management and empowering them to tackle complex real-world analytics tasks. Specifically, our innovative approach includes a scalable synthetic data generation method that enables the creation of a broad spectrum of topics centered on data representation and manipulation. Furthermore, we introduce two new tasks that seamlessly bridge tables and text. We show that such tasks can enhance models' understanding of schema creation and the nuanced translation between natural language and tabular data. Leveraging this data recipe, we post-train a new foundation model, named CoddLLM, based on Mistral-NeMo-12B. To assess the language understanding and reasoning capabilities of LLMs in the realm of data analytics, we contribute AnalyticsMMLU, a benchmark containing thousands of multiple-choice questions on databases, data analysis, and machine learning. Our focus on data discovery, has resulted in the contribution of three comprehensive benchmarks that address both database and data lake scenarios. CoddLLM not only excels in performance but also sets a new standard, achieving the highest average accuracy across eight datasets. It outperforms GPT-3.5-Turbo on AnalyticsMMLU, exceeding GPT-4o by 12.1% in table selection and showing an average improvement of 24.9% in Text-to-SQL compared to the base model.

Since the SciCap datasets launch in 2021, the research community has made significant progress in generating captions for scientific figures in scholarly articles. In 2023, the first SciCap Challenge took place, inviting global teams to use an expanded SciCap dataset to develop models for captioning diverse figure types across various academic fields. At the same time, text generation models advanced quickly, with many powerful pre-trained large multimodal models (LMMs) emerging that showed impressive capabilities in various vision-and-language tasks. This paper presents an overview of the first SciCap Challenge and details the performance of various models on its data, capturing a snapshot of the fields state. We found that professional editors overwhelmingly preferred figure captions generated by GPT-4V over those from all other models and even the original captions written by authors. Following this key finding, we conducted detailed analyses to answer this question: Have advanced LMMs solved the task of generating captions for scientific figures?, Comment: Accepted to TACL 2025

Recent advances have shown that statistical tests for the rank of cross-covariance matrices play an important role in causal discovery. These rank tests include partial correlation tests as special cases and provide further graphical information about latent variables. Existing rank tests typically assume that all the continuous variables can be perfectly measured, and yet, in practice many variables can only be measured after discretization. For example, in psychometric studies, the continuous level of certain personality dimensions of a person can only be measured after being discretized into order-preserving options such as disagree, neutral, and agree. Motivated by this, we propose Mixed data Permutation-based Rank Test (MPRT), which properly controls the statistical errors even when some or all variables are discretized. Theoretically, we establish the exchangeability and estimate the asymptotic null distribution by permutations; as a consequence, MPRT can effectively control the Type I error in the presence of discretization while previous methods cannot. Empirically, our method is validated by extensive experiments on synthetic data and real-world data to demonstrate its effectiveness as well as applicability in causal discovery.

Bug triaging for security vulnerabilities is a critical part of software maintenance, ensuring that the most pressing vulnerabilities are addressed promptly to safeguard system integrity and user data. However, the process is resource-intensive and comes with challenges, including classifying software vulnerabilities, assessing their severity, and managing a high volume of bug reports. In this paper, we present CASEY, a novel approach that leverages Large Language Models (in our case, the GPT model) that automates the identification of Common Weakness Enumerations (CWEs) of security bugs and assesses their severity. CASEY employs prompt engineering techniques and incorporates contextual information at varying levels of granularity to assist in the bug triaging process. We evaluated CASEY using an augmented version of the National Vulnerability Database (NVD), employing quantitative and qualitative metrics to measure its performance across CWE identification, severity assessment, and their combined analysis. CASEY achieved a CWE identification accuracy of 68%, a severity identification accuracy of 73.6%, and a combined accuracy of 51.2% for identifying both. These results demonstrate the potential of LLMs in identifying CWEs and severity levels, streamlining software vulnerability management, and improving the efficiency of security vulnerability triaging workflows., Comment: 16 pages, 22 figures, 6 tables, preprint

Social Cyber Geography is the space in the digital cyber realm that is produced through social relations. Communication in the social media ecosystem happens not only because of human interactions, but is also fueled by algorithmically controlled bot agents. Most studies have not looked at the social cyber geography of bots because they focus on bot activity within a single country. Since creating a bot uses universal programming technology, bots, how prevalent are these bots throughout the world? To quantify bot activity worldwide, we perform a multilingual and geospatial analysis on a large dataset of social data collected from X during the Coronavirus pandemic in 2021. This pandemic affected most of the world, and thus is a common topic of discussion. Our dataset consists of ~100 mil posts generated by ~31mil users. Most bot studies focus only on English-speaking countries, because most bot detection algorithms are built for the English language. However, only 47\% of the bots write in the English language. To accommodate multiple languages in our bot detection algorithm, we built Multilingual BotBuster, a multi-language bot detection algorithm to identify the bots in this diverse dataset. We also create a Geographical Location Identifier to swiftly identify the countries a user affiliates with in his description. Our results show that bots can appear to move from one country to another, but the language they write in remains relatively constant. Bots distribute narratives on distinct topics related to their self-declared country affiliation. Finally, despite the diverse distribution of bot locations around the world, the proportion of bots per country is about 20%. Our work stresses the importance of a united analysis of the cyber and physical realms, where we combine both spheres to inventorize the language and location of social media bots and understand communication strategies.

The first International AI Safety Report comprehensively synthesizes the current evidence on the capabilities, risks, and safety of advanced AI systems. The report was mandated by the nations attending the AI Safety Summit in Bletchley, UK. Thirty nations, the UN, the OECD, and the EU each nominated a representative to the report's Expert Advisory Panel. A total of 100 AI experts contributed, representing diverse perspectives and disciplines. Led by the report's Chair, these independent experts collectively had full discretion over the report's content.

Multimodal Large Language Models (MLLMs) still struggle with hallucinations despite their impressive capabilities. Recent studies have attempted to mitigate this by applying Direct Preference Optimization (DPO) to multimodal scenarios using preference pairs from text-based responses. However, our analysis of representation distributions reveals that multimodal DPO struggles to align image and text representations and to distinguish between hallucinated and non-hallucinated descriptions. To address these challenges, in this work, we propose a Cross-modal Hierarchical Direct Preference Optimization (CHiP) to address these limitations. We introduce a visual preference optimization module within the DPO framework, enabling MLLMs to learn from both textual and visual preferences simultaneously. Furthermore, we propose a hierarchical textual preference optimization module that allows the model to capture preferences at multiple granular levels, including response, segment, and token levels. We evaluate CHiP through both quantitative and qualitative analyses, with results across multiple benchmarks demonstrating its effectiveness in reducing hallucinations. On the Object HalBench dataset, CHiP outperforms DPO in hallucination reduction, achieving improvements of 52.7% and 55.5% relative points based on the base model Muffin and LLaVA models, respectively. We make all our datasets and code publicly available: https://github.com/LVUGAI/CHiP., Comment: Accepted by ICLR 2025