arXiv preprints from January 1, 2026 through September 23, 2026 — 11:04:02 EST
Software engineering (SE) agents powered by large language models are increasingly adopted in practice, yet they often incur substantial monetary cost. We introduce EET, an experience-driven early termination approach that reduces the cost of SE agents while preserving task performance. EET extracts structured experience from prior...
Exposing latent lexical overlap, script romanization has emerged as an effective strategy for improving cross-lingual transfer (XLT) in multilingual language models (mLMs). Most prior work, however, focused on setups that favor romanization the most: (1) transfer from high-resource Latin-script to low-resource non-Latin-script...
A pair of quantized spin-wave modes is driven by two-tone parallel pumping in a YIG microdisk. The nonlinear dynamics is experimentally investigated by probing the resulting steady state, which is found to critically depend on the chosen pair of modes, the detuning between the pump frequencies and the modes parametric resonance, as...
Stellar-mass black holes (BHs) can form from the near-complete collapse of massive stars, causing them to abruptly disappear. The star M31-2014-DS1 in the Andromeda galaxy was reported to exhibit such a disappearance between 2014 and 2022, with properties consistent with the failed explosion of a $\approx 12 - 13$ M$_\odot$ yellow...
This document summarizes the outcomes of the "Science at the Luminosity Frontier: Jefferson Lab at 22 GeV" workshop, held at the INFN Laboratori Nazionali di Frascati in December 2024. The primary goal of the workshop was to critically assess and refine the scientific case for a proposed energy upgrade of the Continuous Electron Beam...
Vulnerabilities severely threaten software systems, making the timely application of security patches crucial for mitigating attacks. However, software vendors often silently patch vulnerabilities with limited disclosure, where Security Patch Detection (SPD) comes to protect software assets. Recently, most SPD studies have targeted...
The algebraic connectivity $a(G)$, defined as the second smallest eigenvalue of the Laplacian matrix $L(G)$, admits a well-known variational characterization involving the minimization of a quadratic form subject to an $\ell_{2}$-norm constraint. In a recent work, Andrade and Dahl (2024) proposed an analogous formulation based on the...
Algorithm extraction aims to synthesize executable programs directly from models trained on algorithmic tasks, enabling de novo recovery of executable mechanisms from weights without relying on human-written target programs. However, applying this paradigm to Transformer is complicated by representation entanglement (e.g.,...
Projectile motion is one of the most fundamental problems in introductory physics, offering a clear context to connect algebraic reasoning with conceptual understanding. This work presents a computational and pedagogical framework that combines the analytical formulation of projectile motion with interactive visualizations developed...
Edge expansion is a parameter indicating how well-connected a graph is. It is useful for designing robust networks, analysing random walks or information flow through a network and is an important notion in theoretical computer science. Modularity is a measure of how well a graph can be partitioned into communities and is widely used...
Analogue quantum simulators offer a promising route to explore quantum many-body dynamics beyond classical reach in the near term. However, their vulnerability to noise limits the accuracy of simulations. Here, we establish a new framework for mitigating noise in analogue quantum simulation, operating in a time-continuous manner. To...
Loop corrections in QED and gravity have recently been conjectured to give rise to an infinite tower of logarithmic soft theorems governing the universal low-energy behavior of photons and gravitons. We explore the implications of this tower for celestial CFT and for the algebra of conformally soft operators. The symmetry-governed...
Generalisation to unseen subjects in EEG-based emotion classification remains a challenge due to high inter-and intra-subject variability. Continual learning (CL) poses a promising solution by learning from a sequence of tasks while mitigating catastrophic forgetting. Regularisation-based CL approaches, such as Elastic Weight...
Polar topological textures in low-dimensional ferroelectrics have emerged as a versatile platform for high-density information storage and neuromorphic computing. While low-order topological states, such as vortices and skyrmions, have been extensively studied, high-order polar topological families remain largely unexplored due to...
Large-scale integration of distributed energy resources has led to a rapid increase in the number of controllable devices and a significant change in system dynamics. This has necessitating the shift towards more distributed and scalable control strategies to manage the increasing system complexity. In this work, we address the...
Context : Ultra Diffuse Galaxies are low surface brightness systems which have been detected in HI in the field, where their line widths sometimes indicate significant dark matter deficits. They are rarely detected in HI in clusters, making their dynamical properties difficult to assess. The relation between field and cluster...
Photonic quantum computers use the bosonic statistics of photons to construct, through quantum interference, the large entangled states required for measurement-based quantum computation. Therefore, any which-way information present in the photons will degrade quantum interference and introduce errors. While quantum error correction...
Active learning (AL) plays a critical role in materials science, enabling applications such as the construction of machine-learning interatomic potentials for atomistic simulations and the operation of self-driving laboratories. Despite its widespread use, the reliability and effectiveness of AL workflows depend on implicit design...
We study the large-scale behaviour of a class of driven diffusive systems modelled by a Stochastic Partial Differential Equation, the Stochastic Burgers Equation (SBE) with general nonlinearity, at the critical dimension and in infinite volume. Our main result shows that, under a logarithmically superdiffusive space-time scaling, it...
The rapid proliferation of synthetic media, presentation attacks, and document forgeries has created significant vulnerabilities in Know Your Customer (KYC) workflows across financial services, telecommunications, and digital-identity ecosystems. Traditional monolithic KYC systems lack the scalability and agility required to counter...
Electron beam energies in Transmission Electron Microscopes (TEMs) reach the relativistic realm constituting Quantum Electrodynamics (QED) the appropriate framework for the study of electron matter interaction in TEMs. We focus on the inelastic scattering of relativistic electrons from a generic oriented target. The inelastic...
The Turing test may or may not be a valid test of machine intelligence. But in an age of generative AI, the test describes the positions we humans occupy. Judging whether or not something is human or machine produced is an everyday condition for many of us, one that involves taking a spectrum of positions along what the essay...
Recent progress in LLM-driven algorithm discovery, exemplified by DeepMind's AlphaEvolve, has produced new best-known solutions for a range of hard geometric and combinatorial problems. This raises a natural question: to what extent can modern off-the-shelf global optimization solvers match such results when the problems are...
The need and theoretical methods for measuring and reducing CO2 emitted by computing systems are well understood, but real-world examples are still limited. We describe a journey towards green software for a live product running on a public cloud. We discuss practical solutions found, in particular using the cost implications of...
Domain-generalized retinal vessel segmentation is critical for automated ophthalmic diagnosis, yet faces significant challenges from domain shift induced by non-uniform illumination and varying contrast, compounded by the difficulty of preserving fine vessel structures. While the Segment Anything Model (SAM) exhibits remarkable...