arXiv preprints from January 1, 2026 through September 11, 2026 — 22:22:20 EST
As Large Language Models (LLMs) become increasingly accessible to end users, an ever-growing number of inference requests are initiated from edge devices and computed on centralized GPU clusters. However, the resulting exponential growth in computation workload is placing significant strain on data centers, while edge devices remain...
This paper presents our experience to understand latency variance caused by kernel and hardware events, which are often invisible at the application level. For this purpose, we have built VarMRI, a tool chain to monitor and analyze those events in the long term. To mitigate the "big data" problem caused by long-term monitoring, VarMRI...
We propose a novel LLM-based framework for reasoning in discrete, game-theoretic tasks, illustrated with \emph{Tic-Tac-Toe}. The method integrates in-context learning with entropy-guided chain-of-thought (CoT) reasoning and adaptive context retrieval. The model dynamically adjusts both the number of retrieved examples and reasoning...
Given the compressed sensing measurements of an unknown vector $z \in \mathbb{R}^n$ using random matrices, we present a simple method to determine $z$ without solving any optimization problem or linear system. Our method uses $Θ(\log n)$ random sensing matrices in $\mathbb{R}^{k \times n}$ and runs in $O(kn\log n)$ time, where $k =...
In this article, we argue that understanding the collective behavior of agents based on large language models (LLMs) is an essential area of inquiry, with important implications in terms of risks and benefits, impacting us as a society at many levels. We claim that the distinctive nature of LLMs--namely, their initialization with...
Entity-level unlearning is usually evaluated by what a model says: whether it stops naming the target, refuses a query, or shifts a Truth Ratio distribution. These output-level tests, however, do not show whether a subject's internal representation has been attenuated. We introduce the Entity Representation Unlearning Framework...
Social networks are typically inferred from indirect observations, such as proximity data; yet, most methods cannot distinguish between absent relationships and actual negative ties, as both can result in few or no interactions. We address the challenge of inferring signed networks from contact patterns while accounting for whether...
In this paper, we introduce monoidal rewriting systems (MRS), an abstraction of string rewriting in which reductions are defined over an arbitrary ambient monoid rather than a free monoid of words. This shift is partly motivated by logic: the class of free monoids is not first-order axiomatizable, so "working in the free setting"...
The research undertakes a comprehensive comparative analysis of Kolmogorov-Arnold Networks (KAN) and Multi-Layer Perceptrons (MLP), highlighting their effectiveness in solving essential computational challenges like nonlinear function approximation, time-series prediction, and multivariate classification. Rooted in Kolmogorov's...
Concept Bottleneck Models (CBMs) improve the explainability of black-box Deep Learning (DL) by introducing intermediate semantic concepts. However, standard CBMs often overlook domain-specific relationships and causal mechanisms, and their dependence on complete concept labels limits applicability in scientific domains where...
Multi-agent systems (MAS) enable complex reasoning by coordinating multiple agents, but often incur high inference latency due to multi-step execution and repeated model invocations, severely limiting their scalability and usability in time-sensitive scenarios. Most existing approaches primarily optimize task performance and inference...
Mobile Ad Hoc Networks (MANETs) are decentralized wireless networks, characterized by their dynamic topologies and node mobility. In the era of cutting-edge technologies, integrating Software-Defined Networking (SDN) with MANETs offers a promising solution to manage these challenges more efficiently. This paper presents a balanced...
The efficacy of autonomous driving systems hinges critically on robust prediction and planning capabilities. However, current benchmarks are impeded by a notable scarcity of scenarios featuring dense traffic, which is essential for understanding and modeling complex interactions among road users. To address this gap, we collaborated...
State-of-the-art video generative models produce promising visual content yet often violate basic physics principles, limiting their utility. While some attribute this deficiency to insufficient physics understanding from pre-training, we find that the shortfall in physics plausibility also stems from suboptimal inference strategies....
Automated perception of urban roadside infrastructure is crucial for smart city management, yet general-purpose models often struggle to capture the necessary fine-grained attributes and domain rules. While Large Vision Language Models (VLMs) excel at open-world recognition, they often struggle to accurately interpret complex facility...
We present a family of open-source Music Foundation Models designed to advance large-scale music understanding and generation across diverse tasks and modalities. Our framework consists of four major components: (1) HeartCLAP, an audio-text alignment model; (2) HeartTranscriptor, a robust lyric recognition model optimized for...
Mobile Ad Hoc Networks (MANETs) and Internet of Things (IoT) networks operate in decentralized and dynamic environments, making them ideal for scenarios lacking traditional infrastructure. However, these networks face challenges such as inefficient routing, limited scalability, and security vulnerabilities due to their decentralized...
Large language models (LLMs) have achieved impressive performance across natural language tasks and are increasingly deployed in real-world applications. Despite extensive safety alignment efforts, recent studies show that such alignment is often shallow and remains vulnerable to jailbreak attacks. Existing defense mechanisms,...
A key challenge in normalizing flows is finding expressive invertible scalar bijections. Existing approaches face trade-offs: affine transformations are smooth and analytically invertible but lack expressivity; monotonic splines offer local control but are only piecewise smooth and act on bounded domains; residual flows achieve...
Certified deletion allows Alice to outsource data to Bob and, at a later time, obtain a verifiable guarantee that the file has been irreversibly deleted at her request. This functionality, while impossible using classical information alone, can be achieved using quantum information. Existing approaches rely either on one-time pad...
The predominance of machine learning models in many spheres of human activity has led to a growing demand for their transparency. The transparency of models makes it possible to discern some factors, such as security or non-discrimination. In this paper, we propose a mixture of transparent local models as an alternative solution for...
Burst errors involving simultaneous insertions, deletions, and substitutions occur in practical scenarios, including DNA data storage and document synchronization, motivating developments of channel codes that can correct such errors. In this paper, we address the problem of constructing error-correcting codes (ECCs) capable of...
Integrated sensing and communication (ISAC) is a cornerstone technology for 6G networks, offering unified support for high-rate communication and high-accuracy sensing. While existing literature extensively covers link-level designs, the transition toward large-scale deployment necessitates a fundamental understanding of network-level...
Images captured in hazy and smoky environments suffer from reduced visibility, posing a challenge when monitoring infrastructures and hindering emergency services during critical situations. The proposed work investigates the use of the deep learning models to enhance the automatic, machine-based readability of gauge in smoky...
The evolution of User Interface design has emphasized the need for efficient, reusable, and editable components to ensure an efficient design process. This research introduces CoGen, a system that uses machine learning techniques to generate reusable UI components directly in Figma, one of the most popular UI design tools. Addressing...