url
stringlengths 33
33
| title
stringlengths 18
214
| date_published
stringdate 2025-03-20 00:07:06
2025-04-17 04:46:57
| abstract
stringlengths 114
1.92k
|
|---|---|---|---|
http://arxiv.org/abs/2504.09378v2 | Can you map it to English? The Role of Cross-Lingual Alignment in Multilingual Performance of LLMs | 2025-04-13T00:01:22+00:00 | Large language models (LLMs) pre-trained predominantly on English text exhibit surprising multilingual capabilities, yet the mechanisms driving cross-lingual generalization remain poorly understood. This work investigates how the alignment of representations for text written in different languages correlates with LLM performance on natural language understanding tasks and translation tasks, both at the language and the instance level. For this purpose, we introduce cross-lingual alignment metrics such as the Discriminative Alignment Index (DALI) to quantify the alignment at an instance level for discriminative tasks. Through experiments on three natural language understanding tasks (Belebele, XStoryCloze, XCOPA), and machine translation, we find that while cross-lingual alignment metrics strongly correlate with task accuracy at the language level, the sample-level alignment often fails to distinguish correct from incorrect predictions, exposing alignment as a necessary but insufficient condition for success. |
http://arxiv.org/abs/2504.09379v1 | Low-Light Image Enhancement using Event-Based Illumination Estimation | 2025-04-13T00:01:33+00:00 | Low-light image enhancement (LLIE) aims to improve the visibility of images captured in poorly lit environments. Prevalent event-based solutions primarily utilize events triggered by motion, i.e., ''motion events'' to strengthen only the edge texture, while leaving the high dynamic range and excellent low-light responsiveness of event cameras largely unexplored. This paper instead opens a new avenue from the perspective of estimating the illumination using ''temporal-mapping'' events, i.e., by converting the timestamps of events triggered by a transmittance modulation into brightness values. The resulting fine-grained illumination cues facilitate a more effective decomposition and enhancement of the reflectance component in low-light images through the proposed Illumination-aided Reflectance Enhancement module. Furthermore, the degradation model of temporal-mapping events under low-light condition is investigated for realistic training data synthesizing. To address the lack of datasets under this regime, we construct a beam-splitter setup and collect EvLowLight dataset that includes images, temporal-mapping events, and motion events. Extensive experiments across 5 synthetic datasets and our real-world EvLowLight dataset substantiate that the devised pipeline, dubbed RetinEV, excels in producing well-illuminated, high dynamic range images, outperforming previous state-of-the-art event-based methods by up to 6.62 dB, while maintaining an efficient inference speed of 35.6 frame-per-second on a 640X480 image. |
http://arxiv.org/abs/2504.09380v1 | Integrated GARCH-GRU in Financial Volatility Forecasting | 2025-04-13T00:04:15+00:00 | In this study, we propose a novel integrated Generalized Autoregressive Conditional Heteroskedasticity-Gated Recurrent Unit (GARCH-GRU) model for financial volatility modeling and forecasting. The model embeds the GARCH(1,1) formulation directly into the GRU cell architecture, yielding a unified recurrent unit that jointly captures both traditional econometric properties and complex temporal dynamics. This hybrid structure leverages the strengths of GARCH in modeling key stylized facts of financial volatility, such as clustering and persistence, while utilizing the GRU's capacity to learn nonlinear dependencies from sequential data. Compared to the GARCH-LSTM counterpart, the GARCH-GRU model demonstrates superior computational efficiency, requiring significantly less training time, while maintaining and improving forecasting accuracy. Empirical evaluation across multiple financial datasets confirms the model's robust outperformance in terms of mean squared error (MSE) and mean absolute error (MAE) relative to a range of benchmarks, including standard neural networks, alternative hybrid architectures, and classical GARCH-type models. As an application, we compute Value-at-Risk (VaR) using the model's volatility forecasts and observe lower violation ratios, further validating the predictive reliability of the proposed framework in practical risk management settings. |
http://arxiv.org/abs/2504.09381v1 | DiTSE: High-Fidelity Generative Speech Enhancement via Latent Diffusion Transformers | 2025-04-13T00:04:48+00:00 | Real-world speech recordings suffer from degradations such as background noise and reverberation. Speech enhancement aims to mitigate these issues by generating clean high-fidelity signals. While recent generative approaches for speech enhancement have shown promising results, they still face two major challenges: (1) content hallucination, where plausible phonemes generated differ from the original utterance; and (2) inconsistency, failing to preserve speaker's identity and paralinguistic features from the input speech. In this work, we introduce DiTSE (Diffusion Transformer for Speech Enhancement), which addresses quality issues of degraded speech in full bandwidth. Our approach employs a latent diffusion transformer model together with robust conditioning features, effectively addressing these challenges while remaining computationally efficient. Experimental results from both subjective and objective evaluations demonstrate that DiTSE achieves state-of-the-art audio quality that, for the first time, matches real studio-quality audio from the DAPS dataset. Furthermore, DiTSE significantly improves the preservation of speaker identity and content fidelity, reducing hallucinations across datasets compared to state-of-the-art enhancers. Audio samples are available at: http://hguimaraes.me/DiTSE |
http://arxiv.org/abs/2504.09382v1 | Modeling Scrap Composition in Electric Arc and Basic Oxygen Furnaces | 2025-04-13T00:07:06+00:00 | This article aims to determine the composition of scrap (recycled material) used in an Electric Arc Furnace (EAF) or basic Oxygen Furnace (BOF) based on the assumption of mass balance. Accurate knowledge of this composition can increase the usage of recycled material to produce steel, reducing the need for raw ore extraction and minimizing environmental impact by conserving natural resources and lowering carbon emissions. The study develops two models to describe the behavior of elements in the EAF or BOF process. A linear state space model is used for elements transferring completely from scrap to steel, while a non-linear state space model is applied to elements moving into both steel and slag. The Kalman filter and unscented Kalman filter are employed to approximate these models, respectively. Importantly, the models leverage only data already collected as part of the standard production process, avoiding the need for additional measurements that are often costly. This article outlines the formulation of both models, the algorithms used, and discusses the hyperparameters involved. We provide practical suggestions on how to choose appropriate hyperparameters based on expert knowledge and historical data. The models are applied to real BOF data. Cu and Cr are chosen as examples for linear and non-linear models, respectively. The results show that both models can reconstruct the composition of scrap for these elements. The findings provide valuable insights for improving process control and ensuring product quality in steelmaking. |
http://arxiv.org/abs/2504.09383v1 | Numerical Calculation of Periods on Schoen's Class of Calabi-Yau Threefolds | 2025-04-13T00:09:54+00:00 | Through classical modularity conjectures, the period integrals of a holomorphic $3$-form on a rigid Calabi-Yau threefold are interesting from the perspective of number theory. Although the (approximate) values of these integrals would be very useful for studying such relations, they are difficult to calculate and generally not known outside of the rare cases in which we can express them exactly. In this paper, we present an efficient numerical method to compute such periods on a wide class of Calabi-Yau threefolds constructed by small resolutions of fiber products of elliptic surfaces over $\mathbf{P}^1$, introduced by C. Schoen in his 1988 paper. Many example results are given, which can easily be calculated with arbitrary precision. We provide tables in which each result is written with precision of 30 decimal places and then compared to period integrals of the appropriate modular form, to confirm accuracy. |
http://arxiv.org/abs/2504.09384v1 | Contour Flow Constraint: Preserving Global Shape Similarity for Deep Learning based Image Segmentation | 2025-04-13T00:34:47+00:00 | For effective image segmentation, it is crucial to employ constraints informed by prior knowledge about the characteristics of the areas to be segmented to yield favorable segmentation outcomes. However, the existing methods have primarily focused on priors of specific properties or shapes, lacking consideration of the general global shape similarity from a Contour Flow (CF) perspective. Furthermore, naturally integrating this contour flow prior image segmentation model into the activation functions of deep convolutional networks through mathematical methods is currently unexplored. In this paper, we establish a concept of global shape similarity based on the premise that two shapes exhibit comparable contours. Furthermore, we mathematically derive a contour flow constraint that ensures the preservation of global shape similarity. We propose two implementations to integrate the constraint with deep neural networks. Firstly, the constraint is converted to a shape loss, which can be seamlessly incorporated into the training phase for any learning-based segmentation framework. Secondly, we add the constraint into a variational segmentation model and derive its iterative schemes for solution. The scheme is then unrolled to get the architecture of the proposed CFSSnet. Validation experiments on diverse datasets are conducted on classic benchmark deep network segmentation models. The results indicate a great improvement in segmentation accuracy and shape similarity for the proposed shape loss, showcasing the general adaptability of the proposed loss term regardless of specific network architectures. CFSSnet shows robustness in segmenting noise-contaminated images, and inherent capability to preserve global shape similarity. |
http://arxiv.org/abs/2504.09385v1 | Expressivity of Quadratic Neural ODEs | 2025-04-13T00:40:17+00:00 | This work focuses on deriving quantitative approximation error bounds for neural ordinary differential equations having at most quadratic nonlinearities in the dynamics. The simple dynamics of this model form demonstrates how expressivity can be derived primarily from iteratively composing many basic elementary operations, versus from the complexity of those elementary operations themselves. Like the analog differential analyzer and universal polynomial DAEs, the expressivity is derived instead primarily from the "depth" of the model. These results contribute to our understanding of what depth specifically imparts to the capabilities of deep learning architectures. |
http://arxiv.org/abs/2504.09386v1 | Unified treatment for in-medium light and heavy clusters with RMF models | 2025-04-13T00:42:02+00:00 | It was shown that light nuclei such as $^4$He, $^8$Be, and $^{12}$C can be well described by RMF models, which enables a unified description for nuclei with baryon numbers $A\gtrsim4$. In this work, we propose a hybrid treatment for investigating the clustering phenomenon in nuclear medium, where clusters ranging from light nuclei (e.g., $^3$H, $^3$He, and $^4$He) to heavy ones (e.g., $^{12}$C, $^{16}$O, $^{40}$Ca, $^{48}$Ca, and $^{208}$Pb) can be treated in a unified manner. In particular, assuming a spherical Wigner-Seitz cell, the clusters are fixed by solving the Dirac equations imposing the Dirichlet-Neumann boundary condition, while the nuclear medium are treated with Thomas-Fermi approximation and take constant densities. In the presence of nuclear medium, the clusters eventually become unbound as density increases, while the root-mean-square charge radii increase. For clusters with different proton and neutron numbers $N_p \neq N_n$, their binding energies varies with the proton fraction of nuclear medium, which are less significant for clusters with $N_p = N_n$. The uncertainties of density functionals on the clustering phenomenon are investigated as well adopting 8 different functionals. Based on the obtained results, an analytical formula describing the binding energies of in-medium clusters is then obtained. The results presented in this work should be useful to understand the clustering phenomenon in both heavy-ion collisions and neutron stars. |
http://arxiv.org/abs/2504.09387v1 | On Language Models' Sensitivity to Suspicious Coincidences | 2025-04-13T00:43:06+00:00 | Humans are sensitive to suspicious coincidences when generalizing inductively over data, as they make assumptions as to how the data was sampled. This results in smaller, more specific hypotheses being favored over more general ones. For instance, when provided the set {Austin, Dallas, Houston}, one is more likely to think that this is sampled from "Texas Cities" over "US Cities" even though both are compatible. Suspicious coincidence is strongly connected to pragmatic reasoning, and can serve as a testbed to analyze systems on their sensitivity towards the communicative goals of the task (i.e., figuring out the true category underlying the data). In this paper, we analyze whether suspicious coincidence effects are reflected in language models' (LMs) behavior. We do so in the context of two domains: 1) the number game, where humans made judgments of whether a number (e.g., 4) fits a list of given numbers (e.g., 16, 32, 2); and 2) by extending the number game setup to prominent cities. For both domains, the data is compatible with multiple hypotheses and we study which hypothesis is most consistent with the models' behavior. On analyzing five models, we do not find strong evidence for suspicious coincidences in LMs' zero-shot behavior. However, when provided access to the hypotheses space via chain-of-thought or explicit prompting, LMs start to show an effect resembling suspicious coincidences, sometimes even showing effects consistent with humans. Our study suggests that inductive reasoning behavior in LMs can be enhanced with explicit access to the hypothesis landscape. |
http://arxiv.org/abs/2504.09388v1 | The Rate-Immediacy Barrier in Explicit Tree Code Constructions | 2025-04-13T00:47:48+00:00 | Since the introduction of tree codes by Schulman (STOC 1993), explicit construction of such codes has remained a notorious challenge. While the construction of asymptotically-good explicit tree codes continues to be elusive, a work by Cohen, Haeupler and Schulman (STOC 2018), as well as the state-of-the-art construction by Ben Yaacov, Cohen, and Yankovitz (STOC 2022) have achieved codes with rate $\Omega(1/\log\log n)$, exponentially improving upon the original construction of Evans, Klugerman and Schulman from 1994. All of these constructions rely, at least in part, on increasingly sophisticated methods of combining (block) error-correcting codes. In this work, we identify a fundamental barrier to constructing tree codes using current techniques. We introduce a key property, which we call immediacy, that, while not required by the original definition of tree codes, is shared by all known constructions and inherently arises from recursive combinations of error-correcting codes. Our main technical contribution is the proof of a rate-immediacy tradeoff, which, in particular, implies that any tree code with constant distance and non-trivial immediacy must necessarily have vanishing rate. By applying our rate-immediacy tradeoff to existing constructions, we establish that their known rate analyses are essentially optimal. More broadly, our work highlights the need for fundamentally new ideas--beyond the recursive use of error-correcting codes--to achieve substantial progress in explicitly constructing asymptotically-good tree codes. |
http://arxiv.org/abs/2504.09389v1 | Beyond Memorization: Mapping the Originality-Quality Frontier of Language Models | 2025-04-13T00:48:58+00:00 | As large language models (LLMs) are increasingly used for ideation and scientific discovery, it is important to evaluate their ability to generate novel output. Prior work evaluates novelty as the originality with respect to training data, but original outputs can be low quality. In contrast, non-expert judges may favor high-quality but memorized outputs, limiting the reliability of human preference as a metric. We propose a new novelty metric for LLM generations that balances originality and quality -- the harmonic mean of the fraction of \ngrams unseen during training and a task-specific quality score. We evaluate the novelty of generations from two families of open-data models (OLMo and Pythia) on three creative tasks: story completion, poetry writing, and creative tool use. We find that LLM generated text is less novel than human written text. To elicit more novel outputs, we experiment with various inference-time methods, which reveals a trade-off between originality and quality. While these methods can boost novelty, they do so by increasing originality at the expense of quality. In contrast, increasing model size or applying post-training reliably shifts the Pareto frontier, highlighting that starting with a stronger base model is a more effective way to improve novelty. |
http://arxiv.org/abs/2504.09390v1 | Numerical simulations of the interaction between the stellar magnetic field and a planet | 2025-04-13T00:53:19+00:00 | Kepler and TESS observations led to the discovery of many close-in super Earths, including some with ultra-short orbital periods ($\lesssim 1$ day). During and shortly after their multi-Myr formation epoch, their GKM host stars generally have kilogauss magnetic fields which can exert torques on the orbits of nearby super- Earths. In this work, we examine one aspect of this interaction: the magnetic torque resulting from Alfv\'en-wing drag on non-corotating, non-magnetized planets engulfed by the host stars' stellar wind. We compute the magnitude of this torque for a range of stellar magnetic field strengths, and planetary orbital velocities. We also model the planets' orbital evolution, taking into account for stellar spin down and magnetic field decay, and derive the boundaries within which ultra-short-period super-Earths can survive. |
http://arxiv.org/abs/2504.09391v1 | Survival of the Optimized: An Evolutionary Approach to T-depth Reduction | 2025-04-13T00:55:18+00:00 | Quantum Error Correction (QEC) is essential for realizing practical Fault-Tolerant Quantum Computing (FTQC) but comes with substantial resource overhead. Quantum circuits must be compiled into the Clifford+T gate set, where the non-transversal nature of the T-gates necessitates costly magic distillation. As circuit complexity grows, so does the T-depth: the sequential T-gate layers, due to the decomposition of arbitrary rotations, further increasing the QEC demands. Optimizing T-depth poses two key challenges: it is NP-hard and existing solutions like greedy or brute-force algorithms are either suboptimal or computationally expensive. We address this by framing the problem as a search task and propose a Genetic Algorithm (GA)-based approach to discover near-optimal T-gate merge patterns across circuit layers. To improve upon convergence and solution quality, we incorporate a mathematical expansion scheme that facilitates reordering layers to identify better merge opportunities, along with a greedy initialization strategy based on T-gate density. Our method achieves up to 79.23% T-depth reduction and 41.86% T-count reduction in large circuits (90-100 qubits). Compared to state-of-the-art methods like the lookahead-based approach, our framework yields an average improvement of 1.2x across varying circuit sizes and T-gate densities. Our approach is hardware-agnostic making it compatible with diverse QEC architectures such as surface codes and QLDPCs, resulting in a scalable and practical optimization framework for near-term fault-tolerant quantum computing. |
http://arxiv.org/abs/2504.09392v1 | Probabilistic Strategies: Definability and the Tensor Completeness Problem | 2025-04-13T01:14:46+00:00 | Programs that combine I/O and countable probabilistic choice, modulo either bisimilarity or trace equivalence, can be seen as describing a probabilistic strategy. For well-founded programs, we might expect to axiomatize bisimilarity via a sum of equational theories and trace equivalence via a tensor of such theories. This is by analogy with similar results for nondeterminism, established previously. While bisimilarity is indeed axiomatized via a sum of theories, and the tensor is indeed at least sound for trace equivalence, completeness in general, remains an open problem. Nevertheless, we show completeness in the case that either the probabilistic choice or the I/O operations used are finitary. We also show completeness up to impersonation, i.e. that the tensor theory regards trace equivalent programs as solving the same system of equations. This entails completeness up to the cancellation law of the probabilistic choice operator. Furthermore, we show that a probabilistic trace strategy arises as the semantics of a well-founded program iff it is victorious. This means that, when the strategy is played against any partial counterstrategy, the probability of play continuing forever is zero. |
http://arxiv.org/abs/2504.09393v1 | Vision Transformers Exhibit Human-Like Biases: Evidence of Orientation and Color Selectivity, Categorical Perception, and Phase Transitions | 2025-04-13T01:15:34+00:00 | This study explored whether Vision Transformers (ViTs) developed orientation and color biases similar to those observed in the human brain. Using synthetic datasets with controlled variations in noise levels, angles, lengths, widths, and colors, we analyzed the behavior of ViTs fine-tuned with LoRA. Our findings revealed four key insights: First, ViTs exhibited an oblique effect showing the lowest angle prediction errors at 180 deg (horizontal) across all conditions. Second, angle prediction errors varied by color. Errors were highest for bluish hues and lowest for yellowish ones. Additionally, clustering analysis of angle prediction errors showed that ViTs grouped colors in a way that aligned with human perceptual categories. In addition to orientation and color biases, we observed phase transition phenomena. While two phase transitions occurred consistently across all conditions, the training loss curves exhibited delayed transitions when color was incorporated as an additional data attribute. Finally, we observed that attention heads in certain layers inherently develop specialized capabilities, functioning as task-agnostic feature extractors regardless of the downstream task. These observations suggest that biases and properties arise primarily from pre-training on the original dataset which shapes the model's foundational representations and the inherent architectural constraints of the vision transformer, rather than being solely determined by downstream data statistics. |
http://arxiv.org/abs/2504.09394v2 | Evaluation Under Imperfect Benchmarks and Ratings: A Case Study in Text Simplification | 2025-04-13T01:36:47+00:00 | Despite the successes of language models, their evaluation remains a daunting challenge for new and existing tasks. We consider the task of text simplification, commonly used to improve information accessibility, where evaluation faces two major challenges. First, the data in existing benchmarks might not reflect the capabilities of current language models on the task, often containing disfluent, incoherent, or simplistic examples. Second, existing human ratings associated with the benchmarks often contain a high degree of disagreement, resulting in inconsistent ratings; nevertheless, existing metrics still have to show higher correlations with these imperfect ratings. As a result, evaluation for the task is not reliable and does not reflect expected trends (e.g., more powerful models being assigned higher scores). We address these challenges for the task of text simplification through three contributions. First, we introduce SynthSimpliEval, a synthetic benchmark for text simplification featuring simplified sentences generated by models of varying sizes. Through a pilot study, we show that human ratings on our benchmark exhibit high inter-annotator agreement and reflect the expected trend: larger models produce higher-quality simplifications. Second, we show that auto-evaluation with a panel of LLM judges (LLMs-as-a-jury) often suffices to obtain consistent ratings for the evaluation of text simplification. Third, we demonstrate that existing learnable metrics for text simplification benefit from training on our LLMs-as-a-jury-rated synthetic data, closing the gap with pure LLMs-as-a-jury for evaluation. Overall, through our case study on text simplification, we show that a reliable evaluation requires higher quality test data, which could be obtained through synthetic data and LLMs-as-a-jury ratings. |
http://arxiv.org/abs/2504.09395v1 | Wavefront Estimation From a Single Measurement: Uniqueness and Algorithms | 2025-04-13T01:38:51+00:00 | Wavefront estimation is an essential component of adaptive optics where the goal is to recover the underlying phase from its Fourier magnitude. While this may sound identical to classical phase retrieval, wavefront estimation faces more strict requirements regarding uniqueness as adaptive optics systems need a unique phase to compensate for the distorted wavefront. Existing real-time wavefront estimation methodologies are dominated by sensing via specialized optical hardware due to their high speed, but they often have a low spatial resolution. A computational method that can perform both fast and accurate wavefront estimation with a single measurement can improve resolution and bring new applications such as real-time passive wavefront estimation, opening the door to a new generation of medical and defense applications. In this paper, we tackle the wavefront estimation problem by observing that the non-uniqueness is related to the geometry of the pupil shape. By analyzing the source of ambiguities and breaking the symmetry, we present a joint optics-algorithm approach by co-designing the shape of the pupil and the reconstruction neural network. Using our proposed lightweight neural network, we demonstrate wavefront estimation of a phase of size $128\times 128$ at $5,200$ frames per second on a CPU computer, achieving an average Strehl ratio up to $0.98$ in the noiseless case. We additionally test our method on real measurements using a spatial light modulator. Code is available at https://pages.github.itap.purdue.edu/StanleyChanGroup/wavefront-estimation/. |
http://arxiv.org/abs/2504.09396v1 | Adaptive Insurance Reserving with CVaR-Constrained Reinforcement Learning under Macroeconomic Regimes | 2025-04-13T01:43:25+00:00 | This paper proposes a reinforcement learning (RL) framework for insurance reserving that integrates tail-risk sensitivity, macroeconomic regime modeling, and regulatory compliance. The reserving problem is formulated as a finite-horizon Markov Decision Process (MDP), in which reserve adjustments are optimized using Proximal Policy Optimization (PPO) subject to Conditional Value-at-Risk (CVaR) constraints. To enhance policy robustness across varying economic conditions, the agent is trained using a regime-aware curriculum that progressively increases volatility exposure. The reward structure penalizes reserve shortfall, capital inefficiency, and solvency floor violations, with design elements informed by Solvency II and Own Risk and Solvency Assessment (ORSA) frameworks. Empirical evaluations on two industry datasets--Workers' Compensation, and Other Liability--demonstrate that the RL-CVaR agent achieves superior performance relative to classical reserving methods across multiple criteria, including tail-risk control (CVaR$_{0.95}$), capital efficiency, and regulatory violation rate. The framework also accommodates fixed-shock stress testing and regime-stratified analysis, providing a principled and extensible approach to reserving under uncertainty. |
http://arxiv.org/abs/2504.09397v1 | Continuous Revival of the Periodic Schrödinger Equation with Piecewise $C^2$ Potential | 2025-04-13T01:43:50+00:00 | In this paper, we investigate the revivals of the one-dimensional periodic Schr\"odinger equation with a piecewise $C^2$ potential function. As has been observed through numerical simulations of the equation with various initial data and potential functions, the solution, while remaining fractalized at irrational times, exhibits a form of revival at rational times. The goal is to prove that the solution at these rational times is given by a finite linear combination of translations and dilations of the initial datum, plus an additional continuous term, which we call "continuous revival". In pursuit of this result, we present a review of relevant properties of the periodic Schr\"odinger equation as an eigenvalue problem, including asymptotic results on both the eigenvalues and eigenfunctions. |
http://arxiv.org/abs/2504.09398v1 | Composable NLP Workflows for BERT-based Ranking and QA System | 2025-04-13T01:48:13+00:00 | There has been a lot of progress towards building NLP models that scale to multiple tasks. However, real-world systems contain multiple components and it is tedious to handle cross-task interaction with varying levels of text granularity. In this work, we built an end-to-end Ranking and Question-Answering (QA) system using Forte, a toolkit that makes composable NLP pipelines. We utilized state-of-the-art deep learning models such as BERT, RoBERTa in our pipeline, evaluated the performance on MS-MARCO and Covid-19 datasets using metrics such as BLUE, MRR, F1 and compared the results of ranking and QA systems with their corresponding benchmark results. The modular nature of our pipeline and low latency of reranker makes it easy to build complex NLP applications easily. |
http://arxiv.org/abs/2504.09399v2 | Rainbow Threshold Graphs | 2025-04-13T01:52:56+00:00 | We define a generalization of threshold graphs which we call $k$-rainbow threshold graphs. We show that the collection of $k$-rainbow threshold graphs do not satisfy the $0$-$1$ law for first order logic and that asymptotically almost surely all $(k+1)$-rainbow threshold graphs are not isomorphic to a $k$-rainbow threshold graph. |
http://arxiv.org/abs/2504.09400v1 | Counting points on some genus zero Shimura curves | 2025-04-13T01:54:34+00:00 | We count certain abelian surfaces with potential quaternionic multiplication defined over a number field $K$ by counting points of bounded height on some genus zero Shimura curves. |
http://arxiv.org/abs/2504.09401v1 | Linear Quadratic Mean Field Stackelberg Games: Open-loop and Feedback Solutions | 2025-04-13T02:08:52+00:00 | This paper investigates open-loop and feedback solutions of linear quadratic mean field (MF) games with a leader and a large number of followers. The leader first gives its strategy and then all the followers cooperate to optimize the social cost as the sum of their costs. By variational analysis with MF approximations, we obtain a set of open-loop controls of players in terms of solutions to MF forward-backward stochastic differential equations (FBSDEs), which is further shown be to an asymptotic Stackelberg equilibrium. By applying the matrix maximum principle, a set of decentralized feedback strategies is constructed for all the players. For open-loop and feedback solutions, the corresponding optimal costs of all players are explicitly given by virtue of the solutions to two Riccati equations, respectively. The performances of two solutions are compared by the numerical simulation. |
http://arxiv.org/abs/2504.09402v1 | Question Tokens Deserve More Attention: Enhancing Large Language Models without Training through Step-by-Step Reading and Question Attention Recalibration | 2025-04-13T02:10:18+00:00 | Large Language Models (LLMs) often struggle with tasks that require a deep understanding of complex questions, especially when faced with long-range dependencies or multi-step reasoning. This work investigates the limitations of current LLMs in question comprehension and identifies three insights: (1) repeating question tokens improves comprehension by increasing attention to question regions, (2) increased backward dependencies negatively affect performance due to unidirectional attentional constraints, and (3) recalibrating attentional mechanisms to prioritize question-relevant regions improves performance. Based on these findings, we first propose a family of prompt-based strategies - Step-by-Step Reading (SSR), SSR+, and SSR++ - that guide LLMs to incrementally process question tokens and align their reasoning with the input structure. These methods significantly improve performance, with SSR++ achieving state-of-the-art results on several benchmarks: 96.66% on GSM8K, 94.61% on ASDiv, and 76.28% on AQuA. Second, we introduce a training-free attention recalibration mechanism that dynamically adjusts attention distributions during inference to emphasize question-relevant regions. This method improves the accuracy of LLaMA 3.1-8B on AQuA by 5.17% without changing model parameters or input prompts. Taken together, our results highlight the importance of structured prompt design and attention optimization in improving LLM comprehension, providing lightweight yet effective tools for improving performance in various NLP tasks. |
http://arxiv.org/abs/2504.09403v1 | Some arithmetic aspects of ortho-integral surfaces | 2025-04-13T02:15:23+00:00 | We investigate ortho-integral (OI) hyperbolic surfaces with totally geodesic boundaries, defined by the property that every orthogeodesic (i.e. a geodesic arc meeting the boundary perpendicularly at both endpoints) has an integer cosh-length. We prove that while only finitely many OI surfaces exist for any fixed topology, infinitely many commensurability classes arise as the topology varies. Moreover, we completely classify OI pants and OI one-holed tori, and show that their doubles are arithmetic surfaces of genus 2 derived from quaternion algebras over $\mathbb{Q}$. |
http://arxiv.org/abs/2504.09404v1 | Nondegenerate Akhmediev breathers and abnormal frequency jumping in multicomponent nonlinear Schrödinger equations | 2025-04-13T02:20:53+00:00 | Nonlinear stage of higher-order modulation instability (MI) phenomena in the frame of multicomponent nonlinear Schr\"odinger equations (NLSEs) are studied analytically and numerically. Our analysis shows that the $N$-component NLSEs can reduce to $N-m+1$ components, when $m(\leq N)$ wavenumbers of the plane wave are equal. As an example, we study systematically the case of three-component NLSEs which cannot reduce to the one- or two-component NLSEs. We demonstrate in both focusing and defocusing regimes, the excitation and existence diagram of a class of nondegenerate Akhmediev breathers formed by nonlinear superposition between several fundamental breathers with the same unstable frequency but corresponding to different eigenvalues. The role of such excitation in higher-order MI is revealed by considering the nonlinear evolution starting with a pair of unstable frequency sidebands. It is shown that the spectrum evolution expands over several higher harmonics and contains several spectral expansion-contraction cycles. In particular, abnormal unstable frequency jumping over the stable gaps between the instability bands are observed in both defocusing and focusing regimes. We outline the initial excitation diagram of abnormal frequency jumping in the frequency-wavenumber plane. We confirm the numerical results by exact solutions of multi-Akhmediev breathers of the multi-component NLSEs. |
http://arxiv.org/abs/2504.09405v1 | Tin-Tin: Towards Tiny Learning on Tiny Devices with Integer-based Neural Network Training | 2025-04-13T02:21:24+00:00 | Recent advancements in machine learning (ML) have enabled its deployment on resource-constrained edge devices, fostering innovative applications such as intelligent environmental sensing. However, these devices, particularly microcontrollers (MCUs), face substantial challenges due to limited memory, computing capabilities, and the absence of dedicated floating-point units (FPUs). These constraints hinder the deployment of complex ML models, especially those requiring lifelong learning capabilities. To address these challenges, we propose Tin-Tin, an integer-based on-device training framework designed specifically for low-power MCUs. Tin-Tin introduces novel integer rescaling techniques to efficiently manage dynamic ranges and facilitate efficient weight updates using integer data types. Unlike existing methods optimized for devices with FPUs, GPUs, or FPGAs, Tin-Tin addresses the unique demands of tiny MCUs, prioritizing energy efficiency and optimized memory utilization. We validate the effectiveness of Tin-Tin through end-to-end application examples on real-world tiny devices, demonstrating its potential to support energy-efficient and sustainable ML applications on edge platforms. |
http://arxiv.org/abs/2504.09406v1 | Stability diagram of layer-polarized quantum Hall states in twisted trilayer graphene | 2025-04-13T02:30:57+00:00 | In the twisted trilayer graphene (tTLG) platform, the rich beating patterns between the three graphene layers give rise to a plethora of new length scales and reconstructed electronic bands arising from the emergent moir\'e and moir\'e-of-moir\'e superlattices. The co-existing lattices and superlattices interact and compete with each other to determine the overall transport properties of tTLG, the hierarchy of which can be electrostatically controlled by tuning the out-of-plane charge distribution or layer polarization. In this work, we measure the stability diagram of layer-polarized quantum Hall states in tTLG by systematically mapping out layer-specific Chern numbers in each layer, and intra- and interlayer Chern transitions as a function of displacement field D and total carrier density n. In contrast to twisted bilayer systems, the rich interplay between the three atomic layers gives rise to a complex layer-polarized stability diagram with unconventional transport features that evolve rapidly with electric and magnetic fields. The stability diagram quantitatively characterizes the interlayer screening and charge distribution in tTLG with implication of strong inter-atomic-layer Coulomb coupling. Our work provides comprehensive guidance and insights into predicting and controlling layer-polarization and interlayer transitions in tTLG, and for tuning the individual role and interactions of each participating constituent towards novel material properties. |
http://arxiv.org/abs/2504.09407v1 | UXAgent: A System for Simulating Usability Testing of Web Design with LLM Agents | 2025-04-13T02:34:22+00:00 | Usability testing is a fundamental research method that user experience (UX) researchers use to evaluate and iterate a web design, but\textbf{ how to evaluate and iterate the usability testing study design } itself? Recent advances in Large Language Model-simulated Agent (\textbf{LLM Agent}) research inspired us to design \textbf{UXAgent} to support UX researchers in evaluating and reiterating their usability testing study design before they conduct the real human-subject study. Our system features a Persona Generator module, an LLM Agent module, and a Universal Browser Connector module to automatically generate thousands of simulated users to interactively test the target website. The system also provides an Agent Interview Interface and a Video Replay Interface so that the UX researchers can easily review and analyze the generated qualitative and quantitative log data. Through a heuristic evaluation, five UX researcher participants praised the innovation of our system but also expressed concerns about the future of LLM Agent usage in UX studies. |
http://arxiv.org/abs/2504.09408v1 | Computationally iterative methods for salt-and-pepper denoising | 2025-04-13T02:39:54+00:00 | Image restoration refers to the process of reconstructing noisy, destroyed, or missing parts of an image, which is an ill-posed inverse problem. A specific regularization term and image degradation are typically assumed to achieve well-posedness. Based on the underlying assumption, an image restoration problem can be modeled as a linear or non-linear optimization problem with or without regularization, which can be solved by iterative methods. In this work, we propose two different iterative methods by linearizing a system of non-linear equations and coupling them with a two-phase iterative framework. The qualitative and quantitative experimental results demonstrate the correctness and efficiency of the proposed methods. |
http://arxiv.org/abs/2504.09409v1 | Bregman Linearized Augmented Lagrangian Method for Nonconvex Constrained Stochastic Zeroth-order Optimization | 2025-04-13T02:44:47+00:00 | In this paper, we study nonconvex constrained stochastic zeroth-order optimization problems, for which we have access to exact information of constraints and noisy function values of the objective. We propose a Bregman linearized augmented Lagrangian method that utilizes stochastic zeroth-order gradient estimators combined with a variance reduction technique. We analyze its oracle complexity, in terms of the total number of stochastic function value evaluations required to achieve an \(\epsilon\)-KKT point in \(\ell_p\)-norm metrics with \(p \ge 2\), where \(p\) is a parameter associated with the selected Bregman distance. In particular, starting from a near-feasible initial point and using Rademacher smoothing, the oracle complexity is in order \(O(p d^{2/p} \epsilon^{-3})\) for \(p \in [2, 2 \ln d]\), and \(O(\ln d \cdot \epsilon^{-3})\) for \(p > 2 \ln d\), where \(d\) denotes the problem dimension. Those results show that the complexity of the proposed method can achieve a dimensional dependency lower than \(O(d)\) without requiring additional assumptions, provided that a Bregman distance is chosen properly. This offers a significant improvement in the high-dimensional setting over existing work, and matches the lowest complexity order with respect to the tolerance \(\epsilon\) reported in the literature. Numerical experiments on constrained Lasso and black-box adversarial attack problems highlight the promising performances of the proposed method. |
http://arxiv.org/abs/2504.09410v1 | Heterogeneous multiscale methods for fourth-order singular perturbations | 2025-04-13T02:46:38+00:00 | We develop a numerical homogenization method for fourth-order singular perturbation problems within the framework of heterogeneous multiscale method. These problems arise from heterogeneous strain gradient elasticity and elasticity models for architectured materials. We establish an error estimate for the homogenized solution applicable to general media and derive an explicit convergence for the locally periodic media with the fine-scale $\varepsilon$. For cell problems of size $\delta=\mathbb{N}\varepsilon$, the classical resonance error $\mathcal{O}(\varepsilon/\delta)$ can be eliminated due to the dominance of the higher-order operator. Despite the occurrence of boundary layer effects, discretization errors do not necessarily deteriorate for general boundary conditions. Numerical simulations corroborate these theoretical findings. |
http://arxiv.org/abs/2504.09411v1 | Hausdorff measure and Fourier dimensions of limsup sets arising in weighted and multiplicative Diophantine approximation | 2025-04-13T02:51:55+00:00 | The classical Khintchine--Jarn\'ik Theorem provides elegant criteria for determining the Lebesgue measure and Hausdorff measure of sets of points approximated by rational points, which has inspired much modern research in metric Diophantine approximation. This paper concerns the Lebesgue measure, Hausdorff measure and Fourier dimension of sets arising in weighted and multiplicative Diophantine approximation. We provide zero-full laws for determining the Lebesgue measure and Hausdorff measure of the sets under consideration. In particular, the criterion for the weighted setup refines a dimensional result given by Li, Liao, Velani, Wang, and Zorin [arXiv: 2410.18578 (2024)], while the criteria for the multiplicative setup answer a question raised by Hussain and Simmons [J. Number Theory (2018)] and extend beyond it. A crucial observation is that, even in higher dimensions, both setups are more appropriately understood as consequences of the `balls-to-rectangles' mass transference principle. We also determine the exact Fourier dimensions of these sets. The result we obtain indicates that, in line with the existence results, these sets are generally non-Salem sets, except in the one-dimensional case. This phenomenon can be partly explained by another result of this paper, which states that the Fourier dimension of the product of two sets equals the minimum of their respective Fourier dimensions. |
http://arxiv.org/abs/2504.09412v1 | Deep Mismatch Channel Estimation in IRS based 6G Communication | 2025-04-13T03:02:45+00:00 | We propose a channel estimation protocol to determine the uplink channel state information (CSI) at the base station for an intelligent reflecting surface (IRS) based wireless communication. More specifically, we develop a channel estimation scheme in a multi-user system with high estimation accuracy and low computational complexity. One of the state-of-the-art approaches to channel estimation is the deep learning-based approach. However, the data-driven model often experiences high computational complexity and, thus, is slow to channel estimation. Inspired by the success of utilizing domain knowledge to build effective data-driven models, the proposed scheme uses the high channel correlation property to train a shallow deep learning model. More specifically, utilizing the one coherent channel estimation, the model predicts the subsequent channel coherence CSI. We evaluate the performance of the proposed scheme in terms of normalized mean square error (NMSE) and spectral efficiency (SE) via simulation. The proposed scheme can estimate the CSI with reasonable success of lower NMSE, higher SE, and lower estimation time than existing schemes. |
http://arxiv.org/abs/2504.09413v1 | Scalable Motion In-betweening via Diffusion and Physics-Based Character Adaptation | 2025-04-13T03:04:25+00:00 | We propose a two-stage framework for motion in-betweening that combines diffusion-based motion generation with physics-based character adaptation. In Stage 1, a character-agnostic diffusion model synthesizes transitions from sparse keyframes on a canonical skeleton, allowing the same model to generalize across diverse characters. In Stage 2, a reinforcement learning-based controller adapts the canonical motion to the target character's morphology and dynamics, correcting artifacts and enhancing stylistic realism. This design supports scalable motion generation across characters with diverse skeletons without retraining the entire model. Experiments on standard benchmarks and stylized characters demonstrate that our method produces physically plausible, style-consistent motions under sparse and long-range constraints. |
http://arxiv.org/abs/2504.09414v1 | Appointed-Time Fault-Tolerant Control for Flexible Hypersonic Vehicles with Unmeasurable States Independent of Initial Errors | 2025-04-13T03:05:57+00:00 | This article aims to derive a practical tracking control algorithm for flexible air-breathing hypersonic vehicles (FAHVs) with lumped disturbances, unmeasurable states and actuator failures. Based on the framework of the backstepping technique, an appointed-time fault-tolerant protocol independent of initial errors is proposed. Firstly, a new type of a state observer is constructed to reconstruct the unmeasurable states. Then, an error transformation function is designed to achieve prescribed performance control that does not depend on the initial tracking error. To deal with the actuator failures, practical fixed-time neural network observers are established to provide the estimation of the lumped disturbances. Finally, the proposed control strategy can ensure the practical fixed-time convergence of the closed-loop system, thereby greatly enhancing the transient performance. The proposed method addresses the challenges of ensuring real-time measurement accuracy for angle of attack and flight path angle in hypersonic vehicles, coupled with potential sudden actuator failures, effectively overcoming the drawback of prescribed performance control that requires knowledge of initial tracking errors. Some simulation results are provided to demonstrate the feasibility and the effectiveness of the proposed strategy |
http://arxiv.org/abs/2504.09415v1 | Nash Equilibrium Between Consumer Electronic Devices and DoS Attacker for Distributed IoT-enabled RSE Systems | 2025-04-13T03:09:47+00:00 | In electronic consumer Internet of Things (IoT), consumer electronic devices as edge devices require less computational overhead and the remote state estimation (RSE) of consumer electronic devices is always at risk of denial-of-service (DoS) attacks. Therefore, the adversarial strategy between consumer electronic devices and DoS attackers is critical. This paper focuses on the adversarial strategy between consumer electronic devices and DoS attackers in IoT-enabled RSE Systems. We first propose a remote joint estimation model for distributed measurements to effectively reduce consumer electronic device workload and minimize data leakage risks. The Kalman filter is deployed on the remote estimator, and the DoS attacks with open-loop as well as closed-loop are considered. We further introduce advanced reinforcement learning techniques, including centralized and distributed Minimax-DQN, to address high-dimensional decision-making challenges in both open-loop and closed-loop scenarios. Especially, the Q-network instead of the Q-table is used in the proposed approaches, which effectively solves the challenge of Q-learning. Moreover, the proposed distributed Minimax-DQN reduces the action space to expedite the search for Nash Equilibrium (NE). The experimental results validate that the proposed model can expeditiously restore the RSE error covariance to a stable state in the presence of DoS attacks, exhibiting notable attack robustness. The proposed centralized and distributed Minimax-DQN effectively resolves the NE in both open and closed-loop case, showcasing remarkable performance in terms of convergence. It reveals that substantial advantages in both efficiency and stability are achieved compared with the state-of-the-art methods. |
http://arxiv.org/abs/2504.09416v1 | Spatially Directional Dual-Attention GAT for Spatial Fluoride Health Risk Modeling | 2025-04-13T03:15:15+00:00 | Environmental exposure to fluoride is a major public health concern, particularly in regions with naturally elevated fluoride concentrations. Accurate modeling of fluoride-related health risks, such as dental fluorosis, requires spatially aware learning frameworks capable of capturing both geographic and semantic heterogeneity. In this work, we propose Spatially Directional Dual-Attention Graph Attention Network (SDD-GAT), a novel spatial graph neural network designed for fine-grained health risk prediction. SDD-GAT introduces a dual-graph architecture that disentangles geographic proximity and attribute similarity, and incorporates a directional attention mechanism that explicitly encodes spatial orientation and distance into the message passing process. To further enhance spatial coherence, we introduce a spatial smoothness regularization term that enforces consistency in predictions across neighboring locations. We evaluate SDD-GAT on a large-scale dataset covering over 50,000 fluoride monitoring samples and fluorosis records across Guizhou Province, China. Results show that SDD-GAT significantly outperforms traditional models and state-of-the-art GNNs in both regression and classification tasks, while also exhibiting improved spatial autocorrelation as measured by Moran's I. Our framework provides a generalizable foundation for spatial health risk modeling and geospatial learning under complex environmental settings. |
http://arxiv.org/abs/2504.09417v1 | Bending-compression coupling in extensible slender microswimmers | 2025-04-13T03:21:05+00:00 | Undulatory slender objects have been a central theme in the hydrodynamics of swimming at low Reynolds number, where the slender body is usually assumed to be inextensible, although some microorganisms and artificial microrobots largely deform with compression and extension. Here, we theoretically study the coupling between the bending and compression/extension shape modes, using a geometrical formulation of microswimmer hydrodynamics to deal with the non-commutative effects between translation and rotation. By means of a coarse-grained minimal model and systematic perturbation expansions for small bending and compression/extension, we analytically derive the swimming velocities and report three main findings. First, we revisit the role of anisotropy in the drag ratio of the resistive force theory and generally demonstrate that no motion is possible for uniform compression with isotropic drag. We then find that the bending-compression/extension coupling generates lateral and rotational motion, which enhances the swimmer's manoeuvrability, as well as changes in progressive velocity at a higher order of expansion, while the coupling effects depend on the phase difference between the two modes. Finally, we demonstrate the importance of often-overlooked Lie bracket contributions in computing net locomotion from a deformation gait. Our study sheds light on compression as a forgotten degree of freedom in swimmer locomotion, with important implications for microswimmer hydrodynamics, including understanding of biological locomotion mechanisms and design of microrobots. |
http://arxiv.org/abs/2504.09418v1 | High stakes exams inflate gender gap thus leading to systematic grading errors in introductory physics | 2025-04-13T03:28:17+00:00 | Previous research has suggested that changing the percentage of the course grade associated with exam grades in STEM courses can change the gender gap in the course. It's also been shown that assessments with the highest stakes have the lowest (relative) scores for female students. Previous research by the authors has shown that the implementation of retake exams can eliminate the gender gap in introductory physics courses. This paper explores several different hypotheses for why retake exams are associated with a zeroed gender gap. Analyzing data from exams with different stakes, we argue that the entire gender gap on introductory physics exams may be due to the stakes associated with those exams. In other words, the data support the idea that a gender grade-gap on exams is not measuring a gender difference in the physics knowledge or physics ability of these students. |
http://arxiv.org/abs/2504.09419v1 | Discovery of a high-velocity cloud of the Milky Way as a potential dark galaxy | 2025-04-13T03:31:53+00:00 | High-velocity clouds (HVCs) are composed of neutral hydrogen (HI) moving at velocities that deviate from the general rotational motion of the Milky Way. Currently, the origins of the HVCs remain poorly known due to the difficulty in determining their distance and the lack of any other suitable identification. Here we report the detection of a compact gas clump in HVC AC-I, which displays characteristics typical of a disk galaxy, named AC G185.0-11.5, using the HI observations. We estimated the distance of AC G185.0-11.5 to be about 277.7 kpc using the Baryonic Tully-Fisher relation and constrained its HI gas mass to be between 3.0*10^7 and 4.7*10^8 solar masses. The distance determination indicates that the HVC AC-I hosting AC G185.0-11.5 is an extragalactic object in the Local Group. The absence of molecular gas and an optical counterpart for AC G185.0-11.5 implies that it may be a rare dark galaxy. |
http://arxiv.org/abs/2504.09420v1 | SaRO: Enhancing LLM Safety through Reasoning-based Alignment | 2025-04-13T03:36:06+00:00 | Current safety alignment techniques for large language models (LLMs) face two key challenges: (1) under-generalization, which leaves models vulnerable to novel jailbreak attacks, and (2) over-alignment, which leads to the excessive refusal of benign instructions. Our preliminary investigation reveals semantic overlap between jailbreak/harmful queries and normal prompts in embedding space, suggesting that more effective safety alignment requires a deeper semantic understanding. This motivates us to incorporate safety-policy-driven reasoning into the alignment process. To this end, we propose the Safety-oriented Reasoning Optimization Framework (SaRO), which consists of two stages: (1) Reasoning-style Warmup (RW) that enables LLMs to internalize long-chain reasoning through supervised fine-tuning, and (2) Safety-oriented Reasoning Process Optimization (SRPO) that promotes safety reflection via direct preference optimization (DPO). Extensive experiments demonstrate the superiority of SaRO over traditional alignment methods. |
http://arxiv.org/abs/2504.09421v2 | ClinicalGPT-R1: Pushing reasoning capability of generalist disease diagnosis with large language model | 2025-04-13T04:00:40+00:00 | Recent advances in reasoning with large language models (LLMs)has shown remarkable reasoning capabilities in domains such as mathematics and coding, yet their application to clinical diagnosis remains underexplored. Here, we introduce ClinicalGPT-R1, a reasoning enhanced generalist large language model for disease diagnosis. Trained on a dataset of 20,000 real-world clinical records, ClinicalGPT-R1 leverages diverse training strategies to enhance diagnostic reasoning. To benchmark performance, we curated MedBench-Hard, a challenging dataset spanning seven major medical specialties and representative diseases. Experimental results demonstrate that ClinicalGPT-R1 outperforms GPT-4o in Chinese diagnostic tasks and achieves comparable performance to GPT-4 in English settings. This comparative study effectively validates the superior performance of ClinicalGPT-R1 in disease diagnosis tasks. Resources are available at https://github.com/medfound/medfound. |
http://arxiv.org/abs/2504.09422v1 | Magnetic Interactions between Nanoscale Domains in Correlated Liquids | 2025-04-13T04:04:11+00:00 | The formation of nanoscale domains (NDs) in correlated liquids and the emerging collective magnetic properties have been suggested as key mechanisms governing ion transport under external magnetic fields (eMFs). However, the molecular-level understanding of these magnetic field-driven phenomena and the interaction between these domains remain elusive. To this end, we introduce a simplified model of a solvated nanoparticle (NP) that consists of localized magnetic domains at their surfaces to represent groups of paramagnetic ions, forming NDs, whose effective magnetic dipole moments are at least one order of magnitude greater than the individual ions. We use classical density functional theory (cDFT) to estimate the effective interactions between these localized magnetic NPs (LMNPs). Our findings indicate that, unlike individual ions, magnetic dipole interactions of NDs in the LMNP model can indeed compete with the electrostatic, van der Waals, and hydration interactions. Depending on the direction of eMF, the cDFT effective interactions between two LMNPs turn out to become more attractive or repulsive, which may play a critical role in ion separation and nucleation processes. This indicates that the cDFT interaction barrier heights can be significantly affected by the magnetic dipole interactions and the barrier heights tend to increase as the size of LMNPs increases. |
http://arxiv.org/abs/2504.09423v1 | Some new Liouville type theorems for 3D steady tropical climate model | 2025-04-13T04:12:03+00:00 | In this paper, we study the Liouville type theorems for the stationary tropical climate model in three dimension. With the help of the delicate estimates of several integrals and an iteration argument, we establish Liouville type theorems under seventeen different assumptions. As a consequence, we show that a smooth solution is trivial provided that they belong to some Lebesgue spaces or satisfy some decay conditions at infinity. Our results extend and improve the recent work of Cho-In-Yang (2024 Appl. Math. Lett. 153 109039). |
http://arxiv.org/abs/2504.09424v1 | Comparing Performance of Preprocessing Techniques for Traffic Sign Recognition Using a HOG-SVM | 2025-04-13T04:13:12+00:00 | This study compares the performance of various preprocessing techniques for Traffic Sign Recognition (TSR) using Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) on the German Traffic Sign Recognition Benchmark (GTSRB) dataset. Techniques such as CLAHE, HUE, and YUV were evaluated for their impact on classification accuracy. Results indicate that YUV in particular significantly enhance the performance of the HOG-SVM classifier (improving accuracy from 89.65% to 91.25%), providing insights into improvements for preprocessing pipeline of TSR applications. |
http://arxiv.org/abs/2504.09425v1 | Optimal Control for Kuramoto Model: from Many-Particle Liouville Equation to Diffusive Mean-Field Problem | 2025-04-13T04:16:38+00:00 | In this paper, we investigate the mean-field optimal control problem of a swarm of Kuramoto oscillators. The controls exploit the self-synchronization property of the oscillators to achieve target density and target phase coherence. In the limit of an infinite number of oscillators the collective dynamics of the agents' density is described by a diffusive mean-field model in the form of a non-local PDE, where the non-locality arises from the synchronization mechanism. We prove the existence of the optimal control of the mean-field model by using $\Gamma$-convergence strategy of the cost functional corresponding to the Liouville equation on the particle level. In the discussion of propagation of chaos for fixed control functions we complete the relative entropy estimate by using large deviation estimate given by \cite{MR3858403}. |
http://arxiv.org/abs/2504.09426v1 | BabyVLM: Data-Efficient Pretraining of VLMs Inspired by Infant Learning | 2025-04-13T04:17:12+00:00 | Human infants rapidly develop visual reasoning skills from minimal input, suggesting that developmentally inspired pretraining could significantly enhance the efficiency of vision-language models (VLMs). Although recent efforts have leveraged infant-inspired datasets like SAYCam, existing evaluation benchmarks remain misaligned--they are either too simplistic, narrowly scoped, or tailored for large-scale pretrained models. Additionally, training exclusively on infant data overlooks the broader, diverse input from which infants naturally learn. To address these limitations, we propose BabyVLM, a novel framework comprising comprehensive in-domain evaluation benchmarks and a synthetic training dataset created via child-directed transformations of existing datasets. We demonstrate that VLMs trained with our synthetic dataset achieve superior performance on BabyVLM tasks compared to models trained solely on SAYCam or general-purpose data of the SAYCam size. BabyVLM thus provides a robust, developmentally aligned evaluation tool and illustrates how compact models trained on carefully curated data can generalize effectively, opening pathways toward data-efficient vision-language learning paradigms. |
http://arxiv.org/abs/2504.09427v1 | Ensemble-Enhanced Graph Autoencoder with GAT and Transformer-Based Encoders for Robust Fault Diagnosis | 2025-04-13T04:21:11+00:00 | Fault classification in industrial machinery is vital for enhancing reliability and reducing downtime, yet it remains challenging due to the variability of vibration patterns across diverse operating conditions. This study introduces a novel graph-based framework for fault classification, converting time-series vibration data from machinery operating at varying horsepower levels into a graph representation. We utilize Shannon's entropy to determine the optimal window size for data segmentation, ensuring each segment captures significant temporal patterns, and employ Dynamic Time Warping (DTW) to define graph edges based on segment similarity. A Graph Auto Encoder (GAE) with a deep graph transformer encoder, decoder, and ensemble classifier is developed to learn latent graph representations and classify faults across various categories. The GAE's performance is evaluated on the Case Western Reserve University (CWRU) dataset, with cross-dataset generalization assessed on the HUST dataset. Results show that GAE achieves a mean F1-score of 0.99 on the CWRU dataset, significantly outperforming baseline models-CNN, LSTM, RNN, GRU, and Bi-LSTM (F1-scores: 0.94-0.97, p < 0.05, Wilcoxon signed-rank test for Bi-LSTM: p < 0.05) -- particularly in challenging classes (e.g., Class 8: 0.99 vs. 0.71 for Bi-LSTM). Visualization of dataset characteristics reveals that datasets with amplified vibration patterns and diverse fault dynamics enhance generalization. This framework provides a robust solution for fault diagnosis under varying conditions, offering insights into dataset impacts on model performance. |
http://arxiv.org/abs/2504.09428v1 | FROG: Effective Friend Recommendation in Online Games via Modality-aware User Preferences | 2025-04-13T04:27:10+00:00 | Due to the convenience of mobile devices, the online games have become an important part for user entertainments in reality, creating a demand for friend recommendation in online games. However, none of existing approaches can effectively incorporate the multi-modal user features (\emph{e.g.}, images and texts) with the structural information in the friendship graph, due to the following limitations: (1) some of them ignore the high-order structural proximity between users, (2) some fail to learn the pairwise relevance between users at modality-specific level, and (3) some cannot capture both the local and global user preferences on different modalities. By addressing these issues, in this paper, we propose an end-to-end model \textsc{FROG} that better models the user preferences on potential friends. Comprehensive experiments on both offline evaluation and online deployment at \kw{Tencent} have demonstrated the superiority of \textsc{FROG} over existing approaches. |
http://arxiv.org/abs/2504.09429v1 | Galois groups of reductions modulo p of D-finite series | 2025-04-13T04:33:30+00:00 | The aim of this paper is to investigate the algebraicity behavior of reductions of $D$-finite power series modulo prime numbers. For many classes of D-finite functions, such as diagonals of multivariate algebraic series or hypergeometric functions, it is known that their reductions modulo prime numbers, when defined, are algebraic. We formulate a conjecture that uniformizes the Galois groups of these reductions across different prime numbers. We then focus on hypergeometric functions, which serves as a test case for our conjecture. Refining the construction of an annihilating polynomial for the reduction of a hypergeometric function modulo a prime number p, we extract information on the respective Galois groups and show that they behave nicely as p varies. |
http://arxiv.org/abs/2504.09430v1 | Predicting ulcer in H&E images of inflammatory bowel disease using domain-knowledge-driven graph neural network | 2025-04-13T04:34:34+00:00 | Inflammatory bowel disease (IBD) involves chronic inflammation of the digestive tract, with treatment options often burdened by adverse effects. Identifying biomarkers for personalized treatment is crucial. While immune cells play a key role in IBD, accurately identifying ulcer regions in whole slide images (WSIs) is essential for characterizing these cells and exploring potential therapeutics. Multiple instance learning (MIL) approaches have advanced WSI analysis but they lack spatial context awareness. In this work, we propose a weakly-supervised model called DomainGCN that employs a graph convolution neural network (GCN) and incorporates domain-specific knowledge of ulcer features, specifically, the presence of epithelium, lymphocytes, and debris for WSI-level ulcer prediction in IBD. We demonstrate that DomainGCN outperforms various state-of-the-art (SOTA) MIL methods and show the added value of domain knowledge. |
http://arxiv.org/abs/2504.09431v1 | Sub-nanosecond in-plane magnetization switching induced by field-like spin-orbit torques from ferromagnets | 2025-04-13T04:46:02+00:00 | Spin-orbit torques (SOTs) generated in SOT-material/ferromagnet structures are classified as damping-like (DL) and field-like (FL) torques for current-driven magnetization switching. It is well known that both DL- and FL-SOTs originate from the SOT-material and DL-SOT dominates the current-driven switching process while FL-SOT contributes limitedly, resulting in an incubation time (several nanoseconds) during collinear magnetization switching with the spin polarization because of the DL attributes. Here we report a FL-SOT originated from the ferromagnet, different from the origin of DL-SOT, and demonstrate that it dominates the collinear magnetization switching. We show that the FL-SOT and resultant collinear switching can be modulated, one order of magnitude and sign reversal, by controlling the ferromagnet. Because of no incubation time and higher charge-to-spin efficiencies in the FL switching, we further show that the switching time can be down to 200 ps with one order lower critical switching current density compared to DL switching. These results indicate that the FL switching may provide a practical solution for magnetic memory in speed-priority cache applications. |
http://arxiv.org/abs/2504.09432v1 | Probing Spin Defects via Single Spin Relaxometry | 2025-04-13T04:47:13+00:00 | Spin defects in solids offer promising platforms for quantum sensing and memory due to their long coherence times and compatibility with quantum networks. Here, we integrate a single nitrogen-vacancy (NV) center in diamond with scanning probe microscopy to discover, read out, and spatially map arbitrary spin-based quantum sensors at the nanoscale. Using the boron vacancy (V$_B^-$) center in hexagonal boron nitride$\unicode{x2013}$an emerging two-dimensional spin system$\unicode{x2013}$as a model, we detect its electron spin resonance through changes in the spin relaxation time ($T_1$) of a nearby NV center, without requiring direct optical excitation or readout of the V$_B^-$ fluorescence. Cross-relaxation between the NV and V$_B^-$ ensembles results in a pronounced NV $T_1$ reduction, enabling nanoscale mapping of spin defect distributions beyond the optical diffraction limit. This approach highlights NV centers as versatile quantum probes for characterizing spin systems, including those emitting at wavelengths beyond the range of silicon-based detectors. Our results open a pathway to hybrid quantum architectures where sensing and readout qubits are decoupled, facilitating the discovery of otherwise inaccessible quantum defects for advanced sensing and quantum networking. |
http://arxiv.org/abs/2504.09433v1 | Strong altermagnetism and topological features in a two-dimensional van der Waals heterostructure via broken time reversal symmetry | 2025-04-13T04:47:42+00:00 | The advent of altermagnetism, a new phase of magnetism, has garnered significant interest due to its extraordinary spin-polarized electronic bands despite zero net magnetization. Such spin-symmetry-guided robust non-relativistic alternating spin splitting in a compensated collinear magnet presents a novel platform for magnetotransport and nontrivial topology. Predominantly, altermagnetic behavior is observed in bulk magnetic materials upon incorporating external perturbations. However, van der Waals heterostructures can offer exceptional flexibility in tailoring their various emerging properties without the need for external perturbations in the two-dimensional regime. Here, an unconventional time reversal symmetry breaking with sizeable spin splitting via broken space-spin symmetry (\textit{P$\mathcal{T}$}) has been demonstrated in an antiferromagnet/nonmagnet vdW heterostructure. The lifted Kramer's degeneracy alongwith spin-orbit interactions result in non-zero Berry curvature, contributing to the outstanding magnetotransport with a large value of anomalous Hall conductivity ($\sim 732.9\ {\mathit{\Omega}}^{-1}{cm}^{-1}$). The presence of relativistic spin-orbit interactions in addition to predominant non-relativistic effects governed spin-momentum locking with a weak out-of-plane Dzyaloshinskii-Moriya interaction, which induces small spin canting. Further, the lowest magnetic anisotropy energy confirms collinear antiferromagnetic ground state. In particular, a nontrivial topology is observed along the surface [001], which is confirmed by the non-zero Chern number. This study provides a novel approach to realize strong altermagnetism in broken space-spin symmetry systems and fosters emergent transport behaviors. |
http://arxiv.org/abs/2504.09434v1 | Constants of motion network revisited | 2025-04-13T04:57:34+00:00 | Discovering constants of motion is meaningful in helping understand the dynamical systems, but inevitably needs proficient mathematical skills and keen analytical capabilities. With the prevalence of deep learning, methods employing neural networks, such as Constant Of Motion nETwork (COMET), are promising in handling this scientific problem. Although the COMET method can produce better predictions on dynamics by exploiting the discovered constants of motion, there is still plenty of room to sharpen it. In this paper, we propose a novel neural network architecture, built using the singular-value-decomposition (SVD) technique, and a two-phase training algorithm to improve the performance of COMET. Extensive experiments show that our approach not only retains the advantages of COMET, such as applying to non-Hamiltonian systems and indicating the number of constants of motion, but also can be more lightweight and noise-robust than COMET. |
http://arxiv.org/abs/2504.09435v1 | Design Probes for AI-Driven AAC: Addressing Complex Communication Needs in Aphasia | 2025-04-13T05:05:12+00:00 | AI offers key advantages such as instant generation, multi-modal support, and personalized adaptability - potential that can address the highly heterogeneous communication barriers faced by people with aphasia (PWAs). We designed AI-enhanced communication tools and used them as design probes to explore how AI's real-time processing and generation capabilities - across text, image, and audio - can align with PWAs' needs in real-time communication and preparation for future conversations respectively. Through a two-phase "Research through Design" approach, eleven PWAs contributed design insights and evaluated four AI-enhanced prototypes. These prototypes aimed to improve communication grounding and conversational agency through visual verification, grammar construction support, error correction, and reduced language processing load. Despite some challenges, such as occasional mismatches with user intent, findings demonstrate how AI's specific capabilities can be advantageous in addressing PWAs' complex needs. Our work contributes design insights for future Augmentative and Alternative Communication (AAC) systems. |
http://arxiv.org/abs/2504.09436v1 | Time-dependent random phase approximation for particle-number fluctuations and correlations in deep-inelastic collisions of $^{144}$Sm+$^{144}$Sm and $^{154}$Sm+$^{154}$Sm | 2025-04-13T05:10:44+00:00 | The fluctuation-dissipation mechanism underlying non-equilibrium transport in low-energy heavy-ion reactions remains unclear. Although the time-dependent Hartree-Fock (TDHF) method provides a reasonable description of average reaction outcomes and one-body dissipation, it is known to significantly underestimate fluctuations of observables. The purpose of this work is to investigate deep-inelastic collisions of 144Sm+144Sm and 154Sm+154Sm with microscopic mean-field approaches and to show a predominant role of one-body dissipation as well as one-body fluctuations and correlation in low-energy heavy-ion reactions. Three dimensional TDHF calculations are carried out for 144Sm+144Sm at Ecm=500 MeV and 154Sm+154Sm at Ecm=485 MeV for a range of impact parameters with Skyrme SLy5 energy density functional. Backward time evolutions are performed as well to evaluate fluctuations and correlation in nucleon numbers within time-dependent random phase approximation (TDRPA). With TDRPA we calculate mass- and charge-number fluctuations, as well as the correlation between neutron and proton transfers, for each impact parameter. We demonstrate that TDRPA quantitatively reproduces the experimental \sigma_{AA}^2-TKEL distributions, whereas it systematically underestimates the charge fluctuation, \sigma_{ZZ}. The double-differential cross sections of reaction products are calculated, showing good agreement with the experimental data. We confirm a long-thought characteristic property that the closed-shell structure limits nucleon transfer at small energy losses, based on our microscopic TDHF and TDRPA calculations. |
http://arxiv.org/abs/2504.09437v1 | PLS-Assisted Offloading for Edge Computing-Enabled Post-Quantum Security in Resource-Constrained Devices | 2025-04-13T05:14:17+00:00 | With the advent of post-quantum cryptography (PQC) standards, it has become imperative for resource-constrained devices (RCDs) in the Internet of Things (IoT) to adopt these quantum-resistant protocols. However, the high computational overhead and the large key sizes associated with PQC make direct deployment on such devices impractical. To address this challenge, we propose an edge computing-enabled PQC framework that leverages a physical-layer security (PLS)-assisted offloading strategy, allowing devices to either offload intensive cryptographic tasks to a post-quantum edge server (PQES) or perform them locally. Furthermore, to ensure data confidentiality within the edge domain, our framework integrates two PLS techniques: offloading RCDs employ wiretap coding to secure data transmission, while non-offloading RCDs serve as friendly jammers by broadcasting artificial noise to disrupt potential eavesdroppers. Accordingly, we co-design the computation offloading and PLS strategy by jointly optimizing the device transmit power, PQES computation resource allocation, and offloading decisions to minimize overall latency under resource constraints. Numerical results demonstrate significant latency reductions compared to baseline schemes, confirming the scalability and efficiency of our approach for secure PQC operations in IoT networks. |
http://arxiv.org/abs/2504.09438v2 | Cartographers in Cubicles: How Training and Preferences of Mapmakers Interplay with Structures and Norms in Not-for-Profit Organizations | 2025-04-13T05:30:09+00:00 | Choropleth maps are a common and effective way to visualize geographic thematic data. Although cartographers have established many principles about map design, data binning and color usage, less is known about how mapmakers make individual decisions in practice. We interview 16 cartographers and geographic information systems (GIS) experts from 13 government organizations, NGOs, and federal agencies about their choropleth mapmaking decisions and workflows. We categorize our findings and report on how mapmakers follow cartographic guidelines and personal rules of thumb, collaborate with other stakeholders within and outside their organization, and how organizational structures and norms are tied to decision-making during data preparation, data analysis, data binning, map styling, and map post-processing. We find several points of variation as well as regularity across mapmakers and organizations and present takeaways to inform cartographic education and practice, including broader implications and opportunities for CSCW, HCI, and information visualization researchers and practitioners. |
http://arxiv.org/abs/2504.09439v1 | Identity-Aware Vision-Language Model for Explainable Face Forgery Detection | 2025-04-13T05:32:40+00:00 | Recent advances in generative artificial intelligence have enabled the creation of highly realistic image forgeries, raising significant concerns about digital media authenticity. While existing detection methods demonstrate promising results on benchmark datasets, they face critical limitations in real-world applications. First, existing detectors typically fail to detect semantic inconsistencies with the person's identity, such as implausible behaviors or incompatible environmental contexts in given images. Second, these methods rely heavily on low-level visual cues, making them effective for known forgeries but less reliable against new or unseen manipulation techniques. To address these challenges, we present a novel personalized vision-language model (VLM) that integrates low-level visual artifact analysis and high-level semantic inconsistency detection. Unlike previous VLM-based methods, our approach avoids resource-intensive supervised fine-tuning that often struggles to preserve distinct identity characteristics. Instead, we employ a lightweight method that dynamically encodes identity-specific information into specialized identifier tokens. This design enables the model to learn distinct identity characteristics while maintaining robust generalization capabilities. We further enhance detection capabilities through a lightweight detection adapter that extracts fine-grained information from shallow features of the vision encoder, preserving critical low-level evidence. Comprehensive experiments demonstrate that our approach achieves 94.25% accuracy and 94.08% F1 score, outperforming both traditional forgery detectors and general VLMs while requiring only 10 extra tokens. |
http://arxiv.org/abs/2504.09440v1 | Enhancing Mathematical Reasoning in Large Language Models with Self-Consistency-Based Hallucination Detection | 2025-04-13T05:47:52+00:00 | Large language models (LLMs) have demonstrated strong mathematical reasoning capabilities but remain susceptible to hallucinations producing plausible yet incorrect statements especially in theorem proving, symbolic manipulation, and numerical computation. While self-consistency (SC) has been explored as a means to improve factuality in LLMs, existing approaches primarily apply SC to final-answer selection, neglecting the logical consistency of intermediate reasoning steps. In this work, we introduce a structured self-consistency framework designed to enhance the reliability of mathematical reasoning. Our method enforces self-consistency across intermediate steps and final outputs, reducing logical inconsistencies and hallucinations. We evaluate our approach across three core mathematical tasks: theorem proving, symbolic transformation, and numerical computation. Experimental results demonstrate that SC significantly improves proof validity, symbolic reasoning accuracy, and numerical stability while maintaining computational efficiency. Further analysis reveals that structured self-consistency not only enhances problem-solving accuracy but also reduces the variance of model-generated outputs. These findings highlight self-consistency as a robust mechanism for improving mathematical reasoning in LLMs, paving the way for more reliable and interpretable AI-driven mathematics. |
http://arxiv.org/abs/2504.09441v1 | Structure-Accurate Medical Image Translation based on Dynamic Frequency Balance and Knowledge Guidance | 2025-04-13T05:48:13+00:00 | Multimodal medical images play a crucial role in the precise and comprehensive clinical diagnosis. Diffusion model is a powerful strategy to synthesize the required medical images. However, existing approaches still suffer from the problem of anatomical structure distortion due to the overfitting of high-frequency information and the weakening of low-frequency information. Thus, we propose a novel method based on dynamic frequency balance and knowledge guidance. Specifically, we first extract the low-frequency and high-frequency components by decomposing the critical features of the model using wavelet transform. Then, a dynamic frequency balance module is designed to adaptively adjust frequency for enhancing global low-frequency features and effective high-frequency details as well as suppressing high-frequency noise. To further overcome the challenges posed by the large differences between different medical modalities, we construct a knowledge-guided mechanism that fuses the prior clinical knowledge from a visual language model with visual features, to facilitate the generation of accurate anatomical structures. Experimental evaluations on multiple datasets show the proposed method achieves significant improvements in qualitative and quantitative assessments, verifying its effectiveness and superiority. |
http://arxiv.org/abs/2504.09442v1 | Photocurrent Nanoscopy of Quantum Hall Bulk | 2025-04-13T05:56:15+00:00 | Understanding nanoscale electronic and thermal transport of two-dimensional (2D) electron systems in the quantum Hall regime, particularly in the bulk insulating state, poses considerable challenges. One of the primary difficulties arises from the presence of chiral edge channels, whose transport behavior obscures the investigation of the insulating bulk. Using near-field (NF) optical and photocurrent (PC) nanoscopy, we probe real-space variations of the optical and thermal dynamics of graphene in the quantum Hall regime without relying on complex sample or electrode geometries. Near the charge neutrality point (CNP), we detect strong optical and photothermal signals from resonant inter-Landau level (LL) magnetoexciton excitations between the 0th and +-1st LLs, which gradually weaken with increasing doping due to Pauli blocking. Interestingly, at higher doping levels and full integer LL fillings, photothermal signals reappear across the entire sample over a ~10-micrometer scale, indicating unexpectedly long cooling lengths and nonlocal photothermal heating through the insulating bulk. This observation suggests thermal conductivity persists for the localized states even as electronic transport is suppressed - a clear violation of the Wiedemann-Franz (WF) law. Our experiments provide novel insights into nanoscale thermal and electronic transport in incompressible 2D gases, highlighting the roles of magnetoexcitons and chiral edge states in the thermo-optoelectric dynamics of Dirac quantum Hall state. |
http://arxiv.org/abs/2504.09443v1 | Deciphering signatures of Kerr-Sen black holes in presence of plasma from the Event Horizon Telescope data | 2025-04-13T06:01:02+00:00 | The present work explores the role of the dilaton charge $r_2$ and the plasma environment in explaining the observed images of M87* and Sgr A*. Dilaton charges are associated with Kerr-Sen black holes, the stationary, axi-symmetric black hole solution in the Einstein-Maxwell-dilaton-axion (EMDA) gravity which arise in the low energy effective action of superstring theories. We investigate the impact of the background spacetime (here dilaton charge and spin) and the plasma environment in modifying the shape and size of the black hole shadow. The theoretically derived shadow is compared with the observed images of M87* and Sgr A* which enables us to constrain the background spacetime in presence of the plasma environment. Our analysis reveals that the shadow of M87* favors the Kerr scenario and rules out $r_2>0.48$, while the shadow of Sgr A* exhibits a marginal preference towards the Kerr-Sen scenario (although GR is allowed within 1-$\sigma$) and rules out $r_2>1$. Thus, large values of dilaton charge are disfavored for M87* and Sgr A* and this result holds good irrespective of the inhomogeneous plasma environment. In fact, the presence of plasma further constrains the allowed parameter space of $r_2$ and within the observed 1-$\sigma$ interval, the present data cannot distinguish between the Kerr and the Kerr-Sen black holes with mild dilaton charges. Moreover, the shadows of M87* and Sgr A* rule out very dense inhomogeneous plasma environments surrounding these objects and hence, black holes with less dense plasma environments seem to be good sites to detect signatures of dilaton charge. These findings not only underscore the importance of considering plasma effects in shadow related studies but also provide a pathway for refining constraints on alternative gravitational theories using black hole observations. |
http://arxiv.org/abs/2504.09444v1 | Dissipation induced localization-delocalization transition in a flat band | 2025-04-13T06:02:16+00:00 | The interplay between dissipation and localization in quantum systems has garnered significant attention due to its potential to manipulate transport properties and induce phase transitions. In this work, we explore the dissipation-induced extended-localized transition in a flat band model, where the system's asymptotic state can be controlled by tailored dissipative operators. By analyzing the steady-state density matrix and dissipative dynamics, we demonstrate that dissipation is able to drive the system to states dominated by either extended or localized modes, irrespective of the initial conditions. The control mechanism relies on the phase properties of the dissipative operators, which selectively favor specific eigenstates of the Hamiltonian. Our findings reveal that dissipation can be harnessed to induce transitions between extended and localized phases, offering a novel approach to manipulate quantum transport in flat band systems. This work not only deepens our understanding of dissipation-induced phenomena in flat band systems but also provides a new avenue for controlling quantum states in open systems. |
http://arxiv.org/abs/2504.09445v1 | New gravitational wave polarization modes in the torsionless spacetime: a general analysis of Palatini theories | 2025-04-13T06:03:01+00:00 | In this study, we investigate the polarization properties of gravitational waves within a torsionless spacetime framework, as described by the Palatini formalism. Our analysis uncovers the presence of two novel polarization modes, referred to as shear modes, which extend beyond the traditional set of six modes in a four-dimensional Riemannian spacetime. These shear modes, uniquely driven by vector degrees of freedom associated with non-metricity, are classified as vector modes, and their detection provides a unique opportunity to explore the fundamental structure of spacetime. Furthermore, we establish a comprehensive theoretical framework within the most general second-order Palatini theory to systematically analyze the polarization modes of gravitational waves, providing robust support for using gravitational wave detectors to test and refine gravitational theories. |
http://arxiv.org/abs/2504.09446v2 | Sparse Deformable Mamba for Hyperspectral Image Classification | 2025-04-13T06:08:19+00:00 | Although Mamba models significantly improve hyperspectral image (HSI) classification, one critical challenge is the difficulty in building the sequence of Mamba tokens efficiently. This paper presents a Sparse Deformable Mamba (SDMamba) approach for enhanced HSI classification, with the following contributions. First, to enhance Mamba sequence, an efficient Sparse Deformable Sequencing (SDS) approach is designed to adaptively learn the ''optimal" sequence, leading to sparse and deformable Mamba sequence with increased detail preservation and decreased computations. Second, to boost spatial-spectral feature learning, based on SDS, a Sparse Deformable Spatial Mamba Module (SDSpaM) and a Sparse Deformable Spectral Mamba Module (SDSpeM) are designed for tailored modeling of the spatial information spectral information. Last, to improve the fusion of SDSpaM and SDSpeM, an attention based feature fusion approach is designed to integrate the outputs of the SDSpaM and SDSpeM. The proposed method is tested on several benchmark datasets with many state-of-the-art approaches, demonstrating that the proposed approach can achieve higher accuracy with less computation, and better detail small-class preservation capability. |
http://arxiv.org/abs/2504.09447v1 | Luttinger compensated magnetic material LaMn2SbO6 | 2025-04-13T06:13:36+00:00 | Unconventional magnetism including altermagnetism and Luttinger compensated magnetism, characterized by its duality of real-space antiferromagnetic alignment and momentum-space spin splitting, has garnered widespread attention. While altermagnetism has been extensively studied, research on Luttinger compensated magnetism remains very rare. In particular, Luttinger com pensated magnetic materials are only theoretically predicted and have not yet been synthesized experimentally. In this study, based on symmetry analysis and the first-principles electronic struc ture calculations, we predict that LaMn2SbO6 is a Luttinger compensated magnetic semiconductor. Given that the Mn ions at opposite spin lattice cannot be connected by any symmetry, the spin splitting in LaMn2SbO6 is isotropic. More importantly, LaMn2SbO6 has already been synthesized experimentally, and its magnetic structure has been confirmed by neutron scattering experiments. Therefore, LaMn2SbO6 serves as an excellent material platform for investigating the novel physical properties of Luttinger compensated magnetic materials. |
http://arxiv.org/abs/2504.09448v1 | InfoBound: A Provable Information-Bounds Inspired Framework for Both OoD Generalization and OoD Detection | 2025-04-13T06:13:37+00:00 | In real-world scenarios, distribution shifts give rise to the importance of two problems: out-of-distribution (OoD) generalization, which focuses on models' generalization ability against covariate shifts (i.e., the changes of environments), and OoD detection, which aims to be aware of semantic shifts (i.e., test-time unseen classes). Real-world testing environments often involve a combination of both covariate and semantic shifts. While numerous methods have been proposed to address these critical issues, only a few works tackled them simultaneously. Moreover, prior works often improve one problem but sacrifice the other. To overcome these limitations, we delve into boosting OoD detection and OoD generalization from the perspective of information theory, which can be easily applied to existing models and different tasks. Building upon the theoretical bounds for mutual information and conditional entropy, we provide a unified approach, composed of Mutual Information Minimization (MI-Min) and Conditional Entropy Maximizing (CE-Max). Extensive experiments and comprehensive evaluations on multi-label image classification and object detection have demonstrated the superiority of our method. It successfully mitigates trade-offs between the two challenges compared to competitive baselines. |
http://arxiv.org/abs/2504.09449v1 | aweSOM: a CPU/GPU-accelerated Self-organizing Map and Statistically Combined Ensemble Framework for Machine-learning Clustering Analysis | 2025-04-13T06:17:35+00:00 | We introduce aweSOM, an open-source Python package for machine learning (ML) clustering and classification, using a Self-organizing Maps (SOM) algorithm that incorporates CPU/GPU acceleration to accommodate large ($N > 10^6$, where $N$ is the number of data points), multidimensional datasets. aweSOM consists of two main modules, one that handles the initialization and training of the SOM, and another that stacks the results of multiple SOM realizations to obtain more statistically robust clusters. Existing Python-based SOM implementations (e.g., POPSOM, Yuan (2018); MiniSom, Vettigli (2018); sklearn-som) primarily serve as proof-of-concept demonstrations, optimized for smaller datasets, but lacking scalability for large, multidimensional data. aweSOM provides a solution for this gap in capability, with good performance scaling up to $\sim 10^8$ individual points, and capable of utilizing multiple features per point. We compare the code performance against the legacy implementations it is based on, and find a 10-100x speed up, as well as significantly improved memory efficiency, due to several built-in optimizations. |
http://arxiv.org/abs/2504.09450v1 | Removable Sets for Fractional Heat and Fractional Bessel-Heat Equations | 2025-04-13T06:19:38+00:00 | We examine the fractional heat diffusion equations $L_{\gamma,a}:=(-\Delta_a)^{\frac{\gamma}{2}}+\partial_t$, where $\Delta_a$ is the Laplace- or the Bessel-Laplace operator. We give conditions for removability which are sufficient and which are necessary, by $L^p$-capacities. Introducing a spherical modulus of smoothness we can treat the Laplace and Bessel-Laplace cases together. |
http://arxiv.org/abs/2504.09451v1 | FractalForensics: Proactive Deepfake Detection and Localization via Fractal Watermarks | 2025-04-13T06:22:23+00:00 | Proactive Deepfake detection via robust watermarks has been raised ever since passive Deepfake detectors encountered challenges in identifying high-quality synthetic images. However, while demonstrating reasonable detection performance, they lack localization functionality and explainability in detection results. Additionally, the unstable robustness of watermarks can significantly affect the detection performance accordingly. In this study, we propose novel fractal watermarks for proactive Deepfake detection and localization, namely FractalForensics. Benefiting from the characteristics of fractals, we devise a parameter-driven watermark generation pipeline that derives fractal-based watermarks and conducts one-way encryption regarding the parameters selected. Subsequently, we propose a semi-fragile watermarking framework for watermark embedding and recovery, trained to be robust against benign image processing operations and fragile when facing Deepfake manipulations in a black-box setting. Meanwhile, we introduce an entry-to-patch strategy that implicitly embeds the watermark matrix entries into image patches at corresponding positions, achieving localization of Deepfake manipulations. Extensive experiments demonstrate satisfactory robustness and fragility of our approach against common image processing operations and Deepfake manipulations, outperforming state-of-the-art semi-fragile watermarking algorithms and passive detectors for Deepfake detection. Furthermore, by highlighting the areas manipulated, our method provides explainability for the proactive Deepfake detection results. |
http://arxiv.org/abs/2504.09452v1 | Stong order 1 adaptive approximation of jump-diffusion SDEs with discontinuous drift | 2025-04-13T06:22:52+00:00 | We present an adaptive approximation scheme for jump-diffusion SDEs with discontinuous drift and (possibly) degenerate diffusion. This transformation-based doubly-adaptive quasi-Milstein scheme is the first scheme that has strong convergence rate $1$ in $L^p$ for $p\in[1,\infty)$ with respect to the average computational cost for these SDEs. To obtain our result, we prove that under slightly stronger assumptions which are still weaker than those in existing literature, a related doubly-adaptive quasi-Milstein scheme has convergence order $1$. This scheme is doubly-adaptive in the sense that it is jump-adapted, i.e.~all jump times of the Poisson noise are grid points, and it includes an adaptive stepsize strategy to account for the discontinuities of the drift. |
http://arxiv.org/abs/2504.09453v1 | Inverse ac Josephson effect in Josephson diode | 2025-04-13T06:27:56+00:00 | We study the Josephson junction where nonreciprocal critical current was induced by the interplay of the $4\pi$-periodic and $2\pi$-periodic current-phase relation of the junction. We take the model of a topological junction which serves as a Josephson diode with nonreciprocal critical currents. For this Josephson diode, we demonstrate an inverse ac Josephson effect where an effective dc voltage is induced by a pure ac driving current. We show that this inverse ac Josephson effect originates from the voltage rectification by the nonreciprocal critical current of the system. We explore the dependence of the induced dc voltage on the amplitude and frequency of the ac driving current and reveal the optimized condition for the inverse ac Josephson effect. |
http://arxiv.org/abs/2504.09454v1 | D$^2$iT: Dynamic Diffusion Transformer for Accurate Image Generation | 2025-04-13T06:33:28+00:00 | Diffusion models are widely recognized for their ability to generate high-fidelity images. Despite the excellent performance and scalability of the Diffusion Transformer (DiT) architecture, it applies fixed compression across different image regions during the diffusion process, disregarding the naturally varying information densities present in these regions. However, large compression leads to limited local realism, while small compression increases computational complexity and compromises global consistency, ultimately impacting the quality of generated images. To address these limitations, we propose dynamically compressing different image regions by recognizing the importance of different regions, and introduce a novel two-stage framework designed to enhance the effectiveness and efficiency of image generation: (1) Dynamic VAE (DVAE) at first stage employs a hierarchical encoder to encode different image regions at different downsampling rates, tailored to their specific information densities, thereby providing more accurate and natural latent codes for the diffusion process. (2) Dynamic Diffusion Transformer (D$^2$iT) at second stage generates images by predicting multi-grained noise, consisting of coarse-grained (less latent code in smooth regions) and fine-grained (more latent codes in detailed regions), through an novel combination of the Dynamic Grain Transformer and the Dynamic Content Transformer. The strategy of combining rough prediction of noise with detailed regions correction achieves a unification of global consistency and local realism. Comprehensive experiments on various generation tasks validate the effectiveness of our approach. Code will be released at https://github.com/jiawn-creator/Dynamic-DiT. |
http://arxiv.org/abs/2504.09455v1 | Enhancing Wide-Angle Image Using Narrow-Angle View of the Same Scene | 2025-04-13T06:36:18+00:00 | A common dilemma while photographing a scene is whether to capture it in wider angle, allowing more of the scene to be covered but in lesser details or to click in narrow angle that captures better details but leaves out portions of the scene. We propose a novel method in this paper that infuses wider shots with finer quality details that is usually associated with an image captured by the primary lens by capturing the same scene using both narrow and wide field of view (FoV) lenses. We do so by training a GAN-based model to learn to extract the visual quality parameters from a narrow angle shot and to transfer these to the corresponding wide-angle image of the scene. We have mentioned in details the proposed technique to isolate the visual essence of an image and to transfer it into another image. We have also elaborately discussed our implementation details and have presented the results of evaluation over several benchmark datasets and comparisons with contemporary advancements in the field. |
http://arxiv.org/abs/2504.09456v1 | Don't Deceive Me: Mitigating Gaslighting through Attention Reallocation in LMMs | 2025-04-13T06:47:32+00:00 | Large Multimodal Models (LMMs) have demonstrated remarkable capabilities across a wide range of tasks. However, their vulnerability to user gaslighting-the deliberate use of misleading or contradictory inputs-raises critical concerns about their reliability in real-world applications. In this paper, we address the novel and challenging issue of mitigating the negative impact of negation-based gaslighting on LMMs, where deceptive user statements lead to significant drops in model accuracy. Specifically, we introduce GasEraser, a training-free approach that reallocates attention weights from misleading textual tokens to semantically salient visual regions. By suppressing the influence of "attention sink" tokens and enhancing focus on visually grounded cues, GasEraser significantly improves LMM robustness without requiring retraining or additional supervision. Extensive experimental results demonstrate that GasEraser is effective across several leading open-source LMMs on the GaslightingBench. Notably, for LLaVA-v1.5-7B, GasEraser reduces the misguidance rate by 48.2%, demonstrating its potential for more trustworthy LMMs. |
http://arxiv.org/abs/2504.09457v1 | Lipschitz regularity of fractional $p$-Laplacian | 2025-04-13T06:52:52+00:00 | In this article, we investigate the H\"{o}lder regularity of the fractional $p$-Laplace equation of the form $(-\Delta_p)^s u=f$ where $p>1, s\in (0, 1)$ and $f\in L^\infty_{\rm loc}(\Omega)$. Specifically, we prove that $u\in C^{0, \gamma_\circ}_{\rm loc}(\Omega)$ for $\gamma_\circ=\min\{1, \frac{sp}{p-1}\}$, provided that $\frac{sp}{p-1}\neq 1$. In particular, it shows that $u$ is locally Lipschitz for $\frac{sp}{p-1}>1$. Moreover, we show that for $\frac{sp}{p-1}=1$, the solution is locally Lipschitz, provided that $f$ is locally H\"{o}lder continuous. Additionally, we discuss further regularity results for the fractional double-phase problems. |
http://arxiv.org/abs/2504.09458v1 | The Whitney method of fundamental solutions with Lusin wavelets | 2025-04-13T07:04:24+00:00 | We establish the theoretical foundation for a variant of the method of fundamental solutions (MFS), where the source points $\{q_j\}_{j=1}^\infty$ accumulate towards the domain in a Whitney fashion, meaning that their separation is proportional to the distance to the domain. We prove that the normalized Lusin wavelets $\psi_j(w) = b_j(w-q_j)^{-2}$ constitute a generalized basis, known as a frame, for the Hardy subspace of $L_2$-traces of holomorphic functions on the domain. Consequently, our method, where $\psi_j$ are used as basis functions in the MFS, enables a numerically stable approximation of solutions to Laplace boundary value problems, even when the solutions lack analytic continuation across the boundary. Despite the source points accumulating towards the domain, our computations show no loss of accuracy near the boundary, in contrast to the boundary integral equation method. |
http://arxiv.org/abs/2504.09459v1 | Measuring Leakage in Concept-Based Methods: An Information Theoretic Approach | 2025-04-13T07:09:55+00:00 | Concept Bottleneck Models (CBMs) aim to enhance interpretability by structuring predictions around human-understandable concepts. However, unintended information leakage, where predictive signals bypass the concept bottleneck, compromises their transparency. This paper introduces an information-theoretic measure to quantify leakage in CBMs, capturing the extent to which concept embeddings encode additional, unintended information beyond the specified concepts. We validate the measure through controlled synthetic experiments, demonstrating its effectiveness in detecting leakage trends across various configurations. Our findings highlight that feature and concept dimensionality significantly influence leakage, and that classifier choice impacts measurement stability, with XGBoost emerging as the most reliable estimator. Additionally, preliminary investigations indicate that the measure exhibits the anticipated behavior when applied to soft joint CBMs, suggesting its reliability in leakage quantification beyond fully synthetic settings. While this study rigorously evaluates the measure in controlled synthetic experiments, future work can extend its application to real-world datasets. |
http://arxiv.org/abs/2504.09460v1 | Imprints of extra dimensions in eccentric EMRI gravitational waveforms | 2025-04-13T07:15:58+00:00 | Studies regarding extra-dimensions have been of great interest in modern theoretical physics, including their observational consequences from future gravitational wave (GW) observatories. In this direction, extreme mass-ratio inspirals (EMRI), attracting considerable interest in GW astronomy and fundamental physics, can potentially provide a useful platform for the search of extra dimensions. In this paper, we examine a rotating braneworld black hole in the context of equatorial eccentric EMRI and attempt to provide an order of magnitude analysis for the extra-dimensional parameter termed "tidal charge". We estimate GW fluxes for the dominant mode and determine the impact of the tidal charge parameter on the orbital evolution. We further evaluate the prospects of detecting such a parameter through mismatch computation. We observe a significant enhancement in the mismatch as the value of orbital eccentricity or tidal charge parameter increases; the phenomenon becomes more obvious for rapidly rotating massive black holes. Thus, the study suggests that eccentric EMRI can potentially probe the existence of extra dimensions with future low-frequency detectors such as the Laser Interferometer Space Antenna (LISA). |
http://arxiv.org/abs/2504.09461v1 | ADDT -- A Digital Twin Framework for Proactive Safety Validation in Autonomous Driving Systems | 2025-04-13T07:17:17+00:00 | Autonomous driving systems continue to face safety-critical failures, often triggered by rare and unpredictable corner cases that evade conventional testing. We present the Autonomous Driving Digital Twin (ADDT) framework, a high-fidelity simulation platform designed to proactively identify hidden faults, evaluate real-time performance, and validate safety before deployment. ADDT combines realistic digital models of driving environments, vehicle dynamics, sensor behavior, and fault conditions to enable scalable, scenario-rich stress-testing under diverse and adverse conditions. It supports adaptive exploration of edge cases using reinforcement-driven techniques, uncovering failure modes that physical road testing often misses. By shifting from reactive debugging to proactive simulation-driven validation, ADDT enables a more rigorous and transparent approach to autonomous vehicle safety engineering. To accelerate adoption and facilitate industry-wide safety improvements, the entire ADDT framework has been released as open-source software, providing developers with an accessible and extensible tool for comprehensive safety testing at scale. |
http://arxiv.org/abs/2504.09462v1 | Arbitrary state creation via controlled measurement | 2025-04-13T07:23:50+00:00 | We propose the algorithm for creating an arbitrary pure quantum superposition state with required accuracy of encoding the amplitudes and phases of this state. The algorithm uses controlled measurement of the ancilla state to avoid the problem of small probability of detecting the required ancilla state. This algorithm can be a subroutine generating the required input state in various algorithms, in particular, in matrix-manipulation algorithms developed earlier. |
http://arxiv.org/abs/2504.09463v1 | Comorbidity-Informed Transfer Learning for Neuro-developmental Disorder Diagnosis | 2025-04-13T07:30:55+00:00 | Neuro-developmental disorders are manifested as dysfunctions in cognition, communication, behaviour and adaptability, and deep learning-based computer-aided diagnosis (CAD) can alleviate the increasingly strained healthcare resources on neuroimaging. However, neuroimaging such as fMRI contains complex spatio-temporal features, which makes the corresponding representations susceptible to a variety of distractions, thus leading to less effective in CAD. For the first time, we present a Comorbidity-Informed Transfer Learning(CITL) framework for diagnosing neuro-developmental disorders using fMRI. In CITL, a new reinforced representation generation network is proposed, which first combines transfer learning with pseudo-labelling to remove interfering patterns from the temporal domain of fMRI and generates new representations using encoder-decoder architecture. The new representations are then trained in an architecturally simple classification network to obtain CAD model. In particular, the framework fully considers the comorbidity mechanisms of neuro-developmental disorders and effectively integrates them with semi-supervised learning and transfer learning, providing new perspectives on interdisciplinary. Experimental results demonstrate that CITL achieves competitive accuracies of 76.32% and 73.15% for detecting autism spectrum disorder and attention deficit hyperactivity disorder, respectively, which outperforms existing related transfer learning work for 7.2% and 0.5% respectively. |
http://arxiv.org/abs/2504.09464v1 | Lim's condition and differentiability | 2025-04-13T07:38:20+00:00 | In their seminal work on quasi-normal structures, Lau and Mah studied weak$^\ast$-normal structure in spaces of operators on a Hilbert space using a geometric property of the dual unit ball called Lim's condition. In this paper, we study a weaker form of Lim's condition for $C^\ast$-algebras and $L^1$-predual spaces, i.e., Banach spaces whose dual is isometric to $L^1(\mu)$ for a positive measure $\mu$. In the case of a separable $C^\ast$-algebra ${\mathcal A}$, we show that this condition implies weak$^*$-normal structure and in the general case, the norm on ${\mathcal A}$ is strongly subdifferentiable. In the case of $L^1$-predual spaces, we show that the condition implies $k$-smoothness of the norm. |
http://arxiv.org/abs/2504.09465v1 | Evolutionary Defense: Advancing Moving Target Strategies with Bio-Inspired Reinforcement Learning to Secure Misconfigured Software Applications | 2025-04-13T07:39:01+00:00 | Improper configurations in software systems often create vulnerabilities, leaving them open to exploitation. Static architectures exacerbate this issue by allowing misconfigurations to persist, providing adversaries with opportunities to exploit them during attacks. To address this challenge, a dynamic proactive defense strategy known as Moving Target Defense (MTD) can be applied. MTD continually changes the attack surface of the system, thwarting potential threats. In the previous research, we developed a proof of concept for a single-player MTD game model called RL-MTD, which utilizes Reinforcement Learning (RL) to generate dynamic secure configurations. While the model exhibited satisfactory performance in generating secure configurations, it grappled with an unoptimized and sparse search space, leading to performance issues. To tackle this obstacle, this paper addresses the search space optimization problem by leveraging two bio-inspired search algorithms: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). Additionally, we extend our base RL-MTD model by integrating these algorithms, resulting in the creation of PSO-RL andGA-RL. We compare the performance of three models: base RL-MTD, GA-RL, and PSO-RL, across four misconfigured SUTs in terms of generating the most secure configuration. Results show that the optimal search space derived from both GA-RL and PSO-RL significantly enhances the performance of the base RL-MTD model compared to the version without optimized search space. While both GA-RL and PSO-RL demonstrate effective search capabilities, PSO-RL slightly outperforms GA-RL for most SUTs. Overall, both algorithms excel in seeking an optimal search space which in turn improves the performance of the model in generating optimal secure configuration. |
http://arxiv.org/abs/2504.09466v1 | AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender | 2025-04-13T07:39:17+00:00 | Despite extensive efforts in safety alignment, large language models (LLMs) remain vulnerable to jailbreak attacks. Activation steering offers a training-free defense method but relies on fixed steering coefficients, resulting in suboptimal protection and increased false rejections of benign inputs. To address this, we propose AdaSteer, an adaptive activation steering method that dynamically adjusts model behavior based on input characteristics. We identify two key properties: Rejection Law (R-Law), which shows that stronger steering is needed for jailbreak inputs opposing the rejection direction, and Harmfulness Law (H-Law), which differentiates adversarial and benign inputs. AdaSteer steers input representations along both the Rejection Direction (RD) and Harmfulness Direction (HD), with adaptive coefficients learned via logistic regression, ensuring robust jailbreak defense while preserving benign input handling. Experiments on LLaMA-3.1, Gemma-2, and Qwen2.5 show that AdaSteer outperforms baseline methods across multiple jailbreak attacks with minimal impact on utility. Our results highlight the potential of interpretable model internals for real-time, flexible safety enforcement in LLMs. |
http://arxiv.org/abs/2504.09467v1 | Micro Heat Engines With Hydrodynamic Flow | 2025-04-13T07:47:25+00:00 | Hydrodynamic flows are often generated in colloidal suspensions. Since colloidal particles are frequently used to construct stochastic heat engines, we study how the hydrodynamic flows influence the output parameters of the engine. We study a single colloidal particle confined in a harmonic trap with time-periodic stiffness that provides the engine protocol, in presence of a steady linear shear flow. The nature of the flow (circular, elliptic or hyperbolic) is externally tunable. At long times, the work done by the flow field is shown to dominate over the thermodynamic (Jarzynski) work done by the trap, if there is an appreciable deviation from the circular flow. The work by the time dependent trap is the sole contributor only for a perfectly circular flow. We also study an extended model, where a microscopic spinning particle (spinor) is tethered close to the colloidal particle, i.e. the working substance of the engine, such that the flow generated by the spinor influences the dynamics of the colloidal particle. We simulate the system and explore the influence of such a flow on the thermodynamics of the engine. We further find that for larger spinning frequencies, the work done by the flow dominates and the system cannot produce thermodynamic work. |
http://arxiv.org/abs/2504.09468v1 | Incubation and Beyond: A Comparative Analysis of ASF Projects Sustainability Impacts on Software Quality | 2025-04-13T07:51:40+00:00 | Free and Open Source Software (FOSS) communities' sustainability, meaning to remain operational without signs of weakening or interruptions to its development, is fundamental for the resilience and continuity of society's digital infrastructure. Many digital services and products either leverage or entirely rely on FOSS in their software stack. FOSS sustainability is a multifaceted concept, and the impact of its decline on community products is less known. In this study, we sought to understand how the different aspects of FOSS sustainability impact software quality from a life-cycle perspective. Specifically, we investigate whether and how support and incubation of FOSS projects or bypassing incubation correlate with software quality outcomes. We selected 342 FOSS projects from the Apache Software Foundation that have either graduated, retired, or bypassed their incubator program. We used 16 sustainability metrics to examine their impact on eight software quality metrics. Using Bayesian data analysis, we found that our selected sustainability metrics exhibit distinct relationships with software quality across different project trajectories. Graduated projects showed the strongest sustainability-software quality (SWQ) relationship, both during and post-incubation. In contrast, retired projects showed weaker relationships, despite receiving similar governance support. Bypassed projects, while not outperforming graduated ones, showed comparable sustainability-SWQ relationships. While structured incubation strengthens sustainability and SWQ in graduated projects, retired projects struggle to maintain strong sustainability-SWQ relationships, indicating that additional factors internal and specific to projects influence sustainability. This effect was evident among bypassed projects; their self-reliant sustainability practices yielded stronger sustainability-SWQ compared to the retired ones. |
http://arxiv.org/abs/2504.09469v1 | Counting ideals in abelian number fields | 2025-04-13T07:55:33+00:00 | Already Dedekind and Weber considered the problem of counting integral ideals of norm at most $x$ in a given number field $K$. Here we improve on the existing results in case $K/\mathbb Q$ is abelian and has degree at least four. For these fields, we obtain as a consequence an improvement of the available results on counting pairs of coprime ideals each having norm at most $x$. |
http://arxiv.org/abs/2504.09470v1 | Spectroscopy of Strange Mesons and First Observation of a Strange Crypto-Exotic State with $J^P=0^-$ | 2025-04-13T07:58:18+00:00 | We measured the strange-meson spectrum in the scattering reaction $K^{-}+p \rightarrow K^{-}\pi^{-}\pi^{-}+p$ with the COMPASS spectrometer at CERN. Using the world's largest sample of this reaction, we performed a comprehensive partial-wave analysis of the mesonic final state. It substantially extends the strange-meson spectrum covering twelve states with masses up to 2.4 GeV/$c^2$. We observe the first candidate for a crypto-exotic strange meson with $J^{P}=0^{-}$ and find $K_3$ and $K_4$ states consistent with predictions for the ground states. |
http://arxiv.org/abs/2504.09471v1 | Optional intervals event, sequential operation and their applications in physics, computer science and applied mathematics | 2025-04-13T08:03:20+00:00 | In this paper, we introduce algebraic theories such as set theory and group theory into the analysis of event execution order. We propose concepts like "optional intervals event" and "sequential operation", summarize their algebraic properties and draw Cayley tables. Based on these efforts, we offer new interpretations for certain physical phenomena and computer application scenarios. Finally, we present other issues derived from this paradigm. These concepts can deepen our understanding of motion and find applications in areas such as event arrangement, physical simulation, and computer modeling |
http://arxiv.org/abs/2504.09472v1 | CamMimic: Zero-Shot Image To Camera Motion Personalized Video Generation Using Diffusion Models | 2025-04-13T08:04:11+00:00 | We introduce CamMimic, an innovative algorithm tailored for dynamic video editing needs. It is designed to seamlessly transfer the camera motion observed in a given reference video onto any scene of the user's choice in a zero-shot manner without requiring any additional data. Our algorithm achieves this using a two-phase strategy by leveraging a text-to-video diffusion model. In the first phase, we develop a multi-concept learning method using a combination of LoRA layers and an orthogonality loss to capture and understand the underlying spatial-temporal characteristics of the reference video as well as the spatial features of the user's desired scene. The second phase proposes a unique homography-based refinement strategy to enhance the temporal and spatial alignment of the generated video. We demonstrate the efficacy of our method through experiments conducted on a dataset containing combinations of diverse scenes and reference videos containing a variety of camera motions. In the absence of an established metric for assessing camera motion transfer between unrelated scenes, we propose CameraScore, a novel metric that utilizes homography representations to measure camera motion similarity between the reference and generated videos. Extensive quantitative and qualitative evaluations demonstrate that our approach generates high-quality, motion-enhanced videos. Additionally, a user study reveals that 70.31% of participants preferred our method for scene preservation, while 90.45% favored it for motion transfer. We hope this work lays the foundation for future advancements in camera motion transfer across different scenes. |
http://arxiv.org/abs/2504.09473v1 | The ANTARES detector: two decades of neutrino searches in the Mediterranean Sea | 2025-04-13T08:06:31+00:00 | Interest for studying cosmic neutrinos using deep-sea detectors has increase after the discovery of a diffuse flux of cosmic neutrinos by the IceCube collaboration and the possibility of wider multi-messenger studies with the observations of gravitational waves. The ANTARES detector was the first neutrino telescope in seawater, operating successfully in the Mediterranean Sea for more than a decade and a half. All challenges related to the operation in the deep sea were accurately addressed by the collaboration. Deployment and connection operations became smoother over time; data taking and constant re-calibration of the detector due to the variable environmental conditions were fully automated. A wealth of results on the subject of astroparticle physics, particle physics and multi-messenger astronomy have been obtained, despite the relative modest size of the detector, paving the way to a new generation of larger undersea detectors. This review summarizes the efforts by the ANTARES collaboration that made the possibility to operate neutrino telescopes in seawater a reality and the results obtained in this endeavor. |
http://arxiv.org/abs/2504.09474v1 | MigGPT: Harnessing Large Language Models for Automated Migration of Out-of-Tree Linux Kernel Patches Across Versions | 2025-04-13T08:08:37+00:00 | Out-of-tree kernel patches are essential for adapting the Linux kernel to new hardware or enabling specific functionalities. Maintaining and updating these patches across different kernel versions demands significant effort from experienced engineers. Large language models (LLMs) have shown remarkable progress across various domains, suggesting their potential for automating out-of-tree kernel patch migration. However, our findings reveal that LLMs, while promising, struggle with incomplete code context understanding and inaccurate migration point identification. In this work, we propose MigGPT, a framework that employs a novel code fingerprint structure to retain code snippet information and incorporates three meticulously designed modules to improve the migration accuracy and efficiency of out-of-tree kernel patches. Furthermore, we establish a robust benchmark using real-world out-of-tree kernel patch projects to evaluate LLM capabilities. Evaluations show that MigGPT significantly outperforms the direct application of vanilla LLMs, achieving an average completion rate of 72.59% (50.74% improvement) for migration tasks. |
http://arxiv.org/abs/2504.09475v1 | Robust Bayesian methods using amortized simulation-based inference | 2025-04-13T08:09:30+00:00 | Bayesian simulation-based inference (SBI) methods are used in statistical models where simulation is feasible but the likelihood is intractable. Standard SBI methods can perform poorly in cases of model misspecification, and there has been much recent work on modified SBI approaches which are robust to misspecified likelihoods. However, less attention has been given to the issue of inappropriate prior specification, which is the focus of this work. In conventional Bayesian modelling, there will often be a wide range of prior distributions consistent with limited prior knowledge expressed by an expert. Choosing a single prior can lead to an inappropriate choice, possibly conflicting with the likelihood information. Robust Bayesian methods, where a class of priors is considered instead of a single prior, can address this issue. For each density in the prior class, a posterior can be computed, and the range of the resulting inferences is informative about posterior sensitivity to the prior imprecision. We consider density ratio classes for the prior and implement robust Bayesian SBI using amortized neural methods developed recently in the literature. We also discuss methods for checking for conflict between a density ratio class of priors and the likelihood, and sequential updating methods for examining conflict between different groups of summary statistics. The methods are illustrated for several simulated and real examples. |
http://arxiv.org/abs/2504.09476v1 | Cloud Uptime Archive: Open-Access Availability Data of Web, Cloud, and Gaming Services | 2025-04-13T08:09:34+00:00 | Cloud services are critical to society. However, their reliability is poorly understood. Towards solving the problem, we propose a standard repository for cloud uptime data. We populate this repository with the data we collect containing failure reports from users and operators of cloud services, web services, and online games. The multiple vantage points help reduce bias from individual users and operators. We compare our new data to existing failure data from the Failure Trace Archive and the Google cluster trace. We analyze the MTBF and MTTR, time patterns, failure severity, user-reported symptoms, and operator-reported symptoms of failures in the data we collect. We observe that high-level user facing services fail less often than low-level infrastructure services, likely due to them using fault-tolerance techniques. We use simulation-based experiments to demonstrate the impact of different failure traces on the performance of checkpointing and retry mechanisms. We release the data, and the analysis and simulation tools, as open-source artifacts available at https://github.com/atlarge-research/cloud-uptime-archive . |
http://arxiv.org/abs/2504.09477v1 | Neighborhood unions and disjoint chorded cycles in $2$-connected graphs | 2025-04-13T08:13:37+00:00 | A {\em chorded cycle} $C$ is a graph that contains a spanning cycle and at least one additional edge. Let $G$ be a graph and let $NC_{2}(G)=min\{|N_{G}(x)\cup N_{G}(y)||xy \notin E(G),x,y \in V(G)\}$. In 2010, Gao \cite{YGJ} and Qiao \cite{SNQ} independently proved if $G$ is a graph of order at least $4s$ and $NC_{2}(G) \geq 4s+1$, then $G$ contains $s$ vertex-disjoint chorded cycles, respectively. In 2022, Gould raised a question of whether increasing connectivity would improve outcome. In this paper, we solved the case that $k=2$, and prove that if $G$ be a $2$-connected graph with at least $4s$ vertices and $NC_{2}(G) \geq 4s$, then $G$ contains $s$ vertex-disjoint chorded cycles. The condition is sharp. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.