arXiv preprints from January 1, 2026 through September 24, 2026 — 01:35:44 EST
Speculative Decoding (SD) accelerates autoregressive large language model (LLM) inference by decoupling generation and verification. While recent methods improve draft quality by tightly coupling the drafter with the target model, the verification mechanism itself remains largely unchanged, relying on strict token-level rejection...
Granular materials subjected to impact loading exhibit highly heterogeneous spatiotemporal dynamics governed by wave propagation, pore collapse, and grain-scale rearrangements. Mesoscale hydrocodes resolve these processes but are computationally expensive, limiting their use in parametric studies and uncertainty quantification. In...
In this paper, we study the timeliness of actions in communication systems where actuation is constrained by control permissions or energy availability. Building on the Age of Actuation (AoA) metric, which quantifies the timeliness of actions independently of data freshness, we introduce a new metric, the \emph{Age of Actuated...
A common solution for mitigating outdated or incorrect information in Large Language Models (LLMs) is to provide updated facts in-context or through knowledge editing. However, these methods introduce knowledge conflicts when the knowledge update fails to overwrite the model's parametric knowledge, which propagate to faulty reasoning....
Generative AI is changing how software is produced and used. In vibe coding, an AI agent builds software by selecting and assembling open-source software (OSS), often without users directly reading documentation, reporting bugs, or otherwise engaging with maintainers. We study the equilibrium effects of vibe coding on the OSS...
Deep Learning (DL) libraries (e.g., PyTorch) are popular in AI development. These libraries are complex and contain bugs. Researchers have proposed various bug-finding techniques for such libraries. Yet, there is much room for improvement. A key challenge in testing DL libraries is the lack of API specifications. Prior testing...
Foundational machine-learned interatomic potentials have emerged as powerful tools for atomistic simulations, promising near first-principles accuracy across diverse chemical spaces at a fraction of the cost of quantum-mechanical calculations. However, the most accurate equivariant architectures rely on Clebsch-Gordan tensor products...
Current alignment-based methods for classification in geometric morphometrics do not generally address the classification of new individuals that were not part of the study sample. However, in the context of infant and child nutritional assessment from body shape images this is a relevant problem. In this setting, classification rules...
Bias in chest X-ray classifiers frequently stems from sex- and age-related shortcuts, leading to systematic underdiagnosis of minority subgroups. Previous pixel-space attribute neutralizers, which rely on convolutional encoders, lessen but do not fully remove this attribute leakage at clinically usable edit strengths. This study...
We construct four-dimensional de Sitter space as an excited state, rather than as a vacuum configuration, in type IIB, heterotic SO(32), and heterotic E_8 \times E_8 string theories. This framework provides a mechanism to evade vacuum-based no-go theorems for de Sitter solutions in string theory. Starting from a generic M-theory...
Large Language Models (LLMs) exhibit social biases, which can lead to harmful stereotypes and unfair outcomes. We propose \textbf{Multi-Persona Thinking (MPT)}, a simple inference-time framework that reduces social bias by encouraging reasoning from multiple perspectives. MPT guides the model to consider contrasting social identities,...
The rapid evolution of Retrieval-Augmented Generation (RAG) toward multimodal, high-stakes enterprise applications has outpaced the development of domain specific evaluation benchmarks. Existing datasets often rely on general-domain corpora or purely textual retrieval, failing to capture the complexity of specialized technical...
The use of artificial intelligence (AI) for drone control can have a transformative impact on drone capabilities, especially when real world information can be integrated with drone sensing, command, and control, part of a growing field of physical AI. Large language models (LLMs) can be advantageous if trained at scale on general...
Federal research funding shapes the direction, diversity, and impact of the US scientific enterprise. Large language models (LLMs) are rapidly diffusing into scientific practice, holding substantial promise while raising widespread concerns. Despite growing attention to AI use in scientific writing and evaluation, little is known...
Online encyclopedias are central to contemporary information infrastructures and have become focal points of debates over ideological bias. Wikipedia, in particular, has long been accused of left-leaning bias, while Grokipedia, an AI-generated encyclopedia launched by xAI, has been framed as a right-leaning alternative. This paper...
This study presents the first large-scale empirical evaluation of quantum machine learning for predicting antibiotic resistance in clinical urine cultures. Antibiotic resistance is amongst the top threats to humanity, and inappropriate antibiotic use is a main driver of resistance. We developed a Quantum Projective Learning (QPL)...
Standard autoregressive decoding in large language models (LLMs) is inherently short-sighted, often failing to find globally optimal reasoning paths due to its token-by-token generation process. While inference-time strategies like foresight sampling attempt to mitigate this by simulating future steps, they typically rely on ad-hoc...
Emergency Department overcrowding is a critical issue that compromises patient safety and operational efficiency, necessitating accurate demand forecasting for effective resource allocation. This study evaluates and compares three distinct predictive models: Seasonal AutoRegressive Integrated Moving Average with eXogenous regressors...
Surface-acoustic-wave (SAW) resonators operating at gigahertz (GHz) frequencies are widely used in wireless telecommunication and quantum information processing. Successful implementation of such resonators calls for detailed microscopic understanding of their mode profiles, energy dissipation channels, and imperfections from...
The safe deployment of large language models (LLMs) in high-stakes fields like biomedicine, requires them to be able to reason about cause and effect. We investigate this ability by testing 13 open-source LLMs on a fundamental task: pairwise causal discovery (PCD) from text. Our benchmark, using 12 diverse datasets, evaluates two core...
We study multi-agent contract design, where a principal incentivizes a team of agents to take costly actions that jointly determine the project success via a combinatorial reward function. While prior work largely focuses on unconstrained contracts that allow heterogeneous payments across agents, many real-world environments limit...
Biological membranes are dynamic surfaces whose shape and function are critically influenced by protein inclusions (PIs). While membrane deformations induced by PIs have been extensively studied in the small-deformation regime, a variety of processes involves strong membrane deformations. We investigate the interaction between lipid...
This paper examines how to make large language models reliable for high-stakes legal work by reducing hallucinations. It distinguishes three AI paradigms: (1) standalone generative models ("creative oracle"), (2) basic retrieval-augmented systems ("expert archivist"), and (3) an advanced, end-to-end optimized RAG system ("rigorous...
Novel view synthesis from low dynamic range (LDR) blurry images, which are common in the wild, struggles to recover high dynamic range (HDR) and sharp 3D representations in extreme lighting conditions. Although existing methods employ event data to address this issue, they ignore the sensor-physics mismatches between the camera output...
Graph neural network (GNN) have demonstrated exceptional performance in solving critical problems across diverse domains yet remain susceptible to backdoor attacks. Existing studies on backdoor attack for graph classification are limited to single target attack using subgraph replacement based mechanism where the attacker implants...