arXiv preprints from January 1, 2026 through September 28, 2026 — 14:43:14 EST
We present results from a seven-season (2018-2024) monitoring campaign of the gravitationally lensed quasar system PS J2305+3714 using the 1.5-m Maidanak Telescope in the optical R-band. From these data, the amplitude of possible microlensing variability does not exceed 10 mmag on a 7-yr timescale. Additionally, we measure a time...
Accurate fMRI analysis requires sensitivity to temporal structure across multiple scales, as BOLD signals encode cognitive processes that emerge from fast transient dynamics to slower, large-scale fluctuations. Existing deep learning (DL) approaches to temporal modeling face challenges in jointly capturing these dynamics over long...
Solving image reconstruction problems of the form \(\mathbf{A} \mathbf{x} = \mathbf{y}\) remains challenging due to ill-posedness and the lack of large-scale supervised datasets. Deep Equilibrium (DEQ) models have been used successfully but typically require supervised pairs \((\mathbf{x},\mathbf{y})\). In many practical settings,...
Language learning is a multifaceted process. Insufficient vocabulary can hinder communication and lead to demotivation. For non-Chinese speaking (NCS) students, learning Traditional Chinese (Cantonese) poses distinct challenges, particularly due to the complexity of converting spoken and written forms. To address this issue, this...
Existing user simulation approaches focus on generating user-like responses in dialogue. They often assume that the provided persona is sufficient for producing such responses, without verifying whether critical personas are supplied. This raises concerns about the validity of simulation results. To address this issue, we study the...
Given natural parameters s and r, where $2\leq s\leq r$, we consider the distribution of a random variable $ξ=\sum\limits_{k=1}^{\infty}s^{-k}ξ_k\equivΔ^{r_s}_{ξ_1ξ_2...ξ_k...},$ where $(ξ_k)$ is a sequence of independent random variables taking values in $\{0,1,...,r\}$ with probabilities $p_0,p_1,...,p_r$, respectively, and all $...
This work addresses critical challenges to academic integrity, including plagiarism, fabrication, and verification of authorship of educational content, by proposing a Natural Language Processing (NLP)-based framework for authenticating students' content through author attribution and style change detection. Despite some initial...
Slot Attention (SA) with pretrained diffusion models has recently shown promise for object-centric learning (OCL), but suffers from slot entanglement and weak alignment between object slots and image content. We propose Contrastive Object-centric Diffusion Alignment (CODA), a simple extension that (i) employs register slots to absorb...
In-space manufacturing is essential for achieving long-term planetary colonization, particularly on Mars, where material transport from Earth is both costly and logistically restrictive. Traditional subtractive manufacturing methods are highly equipment-, energy-, and material-intensive, making additive manufacturing (AM) a more...
Clinical decision-making demands uncertainty quantification that provides both distribution-free coverage guarantees and risk-adaptive precision, requirements that existing methods fail to jointly satisfy. We present a hybrid Bayesian-conformal framework that addresses this fundamental limitation in healthcare predictions. Our...
We present UniSH, a unified, feed-forward framework for joint metric-scale 3D scene and human reconstruction. A key challenge in this domain is the scarcity of large-scale, annotated real-world data, forcing a reliance on synthetic datasets. This reliance introduces a significant sim-to-real domain gap, leading to poor generalization,...
Artificial intelligence has revolutionized optical device design, overcoming the efficiency bottlenecks of traditional methods. For holographic metasurfaces, conventional iterative algorithms suffer from time-consuming iterations and convergence stagnation, especially as the complexity of 3D target fields increases. While recent...
We present the first search for new flavor-dependent long-range interactions (LRI) of neutrinos using publicly available 8 years of high-purity $ν_μ$ CC data from IceCube DeepCore. These interactions are mediated by ultra-light gauge bosons with masses below $10^{-10}$ eV, which can arise due to a new lepton-number gauge symmetry,...
Understanding individual-level human mobility is critical for a wide range of applications. As such, real-world trajectory datasets provide valuable insights into actual movement behaviors and patterns of life but are often constrained by data sparsity and participant bias. Synthetic data, by contrast, offers scalability and...
Creativity is increasingly recognized as an important skill in education, and storytelling can enhance motivation and engagement among students. However, conventional storytelling methods often lack the interactive elements necessary to engage students. To this end, this study examines the impact of an interactive digital storytelling...
Open questions on the fundamental nature of the strong force endure and the Large Hadron Collider (LHC) is a once-in-a-generation laboratory elucidating its quantum origins. This document summarizes the plenary overview talk titled "QCD Studies at the LHC" presented at the Lepton Photon Symposium 2025. Selected results highlight...
We introduce an operator-theoretic framework for analyzing directional dependence in multivariate time series based on order-constrained spectral non-invariance. Directional influence is defined as the sensitivity of second-order dependence operators to admissible, order-preserving temporal deformations of a designated source...
We introduce AceFF, a pre-trained machine learning interatomic potential (MLIP) optimized for small molecule drug discovery. While MLIPs have emerged as efficient alternatives to Density Functional Theory (DFT), generalizability across diverse chemical spaces remains difficult. AceFF addresses this via a refined TensorNet2...
Multi-robot systems are widely used for coverage tasks that require efficient coordination across large environments. In Multi-Robot Coverage Path Planning (MCPP), the objective is typically to minimize the makespan by generating non-overlapping paths for full-area coverage. However, most existing methods assume uniform importance...
Large language models (LLMs) have been widely used for mental health support. However, current safety evaluations in this field are mostly limited to detecting whether LLMs output prohibited words in single-turn conversations, neglecting the gradual erosion of safety boundaries in long dialogues. Examples include making definitive...
The fast integration of artificial intelligence into mobile applications has completely changed the digital landscape; however, the impact of this change on user perception of AI features remains poorly understood. This large-scale analysis examined 1,484,633 mobile application reviews across 422 applications (200 AI-featuring, 222...
To enhance the reproducibility and reliability of deep learning models, we address a critical gap in current training methodologies: the lack of mechanisms that ensure consistent and robust performance across runs. Our empirical analysis reveals that even under controlled initialization and training conditions, the accuracy of the...
Over the past decade, numerous theories have been proposed to explain the widespread vulnerability of deep neural networks to adversarial evasion attacks. Among these, the theory of non-robust features proposed by Ilyas et al. has been widely accepted, showing that brittle but predictive features of the data distribution can be...
Light harvesting 2 (LH2) complex is a primary component of the photosynthetic unit of purple bacteria that is responsible for harvesting and relaying excitons. The electronic absorption line shape of LH2 contains two major bands at 800 nm and 850 nm wavelength regions. Under low light condition, some species of purple bacteria replace...
Large language models (LLMs) have demonstrated significant advancements in reasoning and code generation, but efficiently creating new benchmarks to evaluate these capabilities remains a challenge. Traditional benchmark creation relies on manual human effort, which is expensive and time-consuming. Furthermore, existing benchmarks...