arXiv preprints from January 1, 2026 through September 23, 2026 — 13:57:34 EST
While Large Language Model (LLM) role-playing agents have advanced rapidly, it remains unclear which profile elements genuinely drive role-playing quality. To bridge this gap, we introduce a systematic diagnostic framework that disentangles the impact of character profiles along three axes: Familiarity (Known vs. Unknown), Structure...
The rapid advancement of AI-generated content (AIGC) has escalated the threat of deepfakes, from facial manipulations to the synthesis of entire photorealistic human bodies. However, existing detection methods remain fragmented, specializing either in facial-region forgeries or full-body synthetic images, and consequently fail to...
With the rapid advancement of large language models (LLMs) technologies, their application in the domain of autonomous driving has become increasingly widespread. However, existing methods suffer from unstructured reasoning, poor generalization, and misalignment with human driving intent. While Chain-of-Thought (CoT) reasoning...
The deep sea is weakly stratified in density but shows considerable variations in turbulent motions in all three directions. When registered by moored high-resolution temperature 'T'-sensors, the motions cause variations of 0.01degrC or less and in time of minutes or less, which is much faster than hours or longer of internal waves....
PdTe has recently been reported to be a type-II Dirac semimetal while a bulk nodal and surface nodeless superconductivity (SC) has been claimed to coexist. In this work, we applied point-contact spectroscopy (PCS) method to systematically study the superconducting gap in PdTe single crystals with a SC transition temperature...
The reliability of large language models (LLMs) in production environments remains significantly constrained by their propensity to generate hallucinations -- fluent, plausible-sounding outputs that contradict or fabricate information. While hallucination detection has recently emerged as a priority in English-centric benchmarks,...
Fine-tuning large language models (LLMs) achieves strong performance but is often limited by the memory overhead of backpropagation. Zeroth-order (ZO) optimization avoids this overhead by estimating gradients through forward passes alone, yet it typically converges slowly because random Gaussian perturbations yield high-variance...
Root cause analysis in modern cloud infrastructure demands sophisticated understanding of heterogeneous data sources, particularly time-series performance metrics that involve core failure signatures. While large language models demonstrate remarkable capabilities in textual reasoning, their discrete token-based architecture creates...
We compute numerically the $L^2$ Marcinkiewicz-Zygmund constants of cubature rules, with a special attention to their role in polynomial approximation by orthogonal bases. We test some relevant rules on domains such as the interval, the square, the disk, the triangle, the cube and the sphere. The approximation power of the...
Graph Neural Networks (GNNs) are powerful tools for learning graph-structured data, but their scalability is hindered by inefficient mini-batch generation, data transfer bottlenecks, and costly inter-GPU synchronization. Existing training frameworks fail to overlap these stages, leading to suboptimal resource utilization. This paper...
Integrating image generation and understanding into a single framework has become a pivotal goal in the multimodal domain. However, how understanding can effectively assist generation has not been fully explored. Unlike previous works that focus on leveraging reasoning abilities and world knowledge from understanding models, this...
Rapid e-commerce growth has pushed last-mile delivery networks to their limits, where small routing gains translate into lower costs, faster service, and fewer emissions. Classical heuristics struggle to adapt when travel times are highly asymmetric (e.g., one-way streets, congestion). A deep learning-based approach to the last-mile...
Non-volatile Neuromorphic Computing (NC) elements utilizing Spin Orbit Torque (SOT) provide a viable solution to alleviate the memory wall bottleneck in contemporary computing systems. However, the two challenges, low SOT efficiency and the need for in plane symmetry breaking field for perpendicular magnetization switching, greatly...
In this study, we introduce the theory of what we call Hecke vector-forms. A Hecke vector-form can be viewed as a vector function representation of some quasiautomorphic form that transforms like an automorphic form on an arbitrarily chosen Hecke triangle group. In other words, because quasiautomorphic forms have complicated...
Agentic search has emerged as a promising paradigm for complex information seeking by enabling Large Language Models (LLMs) to interleave reasoning with tool use. However, prevailing systems rely on monolithic agents that suffer from structural bottlenecks, including unconstrained reasoning outputs that inflate trajectories, sparse...
High-dimensional dynamical systems of interacting degrees of freedom are ubiquitous in the study of complex systems. When the directed interactions are totally uncorrelated, sufficiently strong and non-linear, many of these systems exhibit a chaotic attractor characterized by a positive maximal Lyapunov exponent (MLE). On the...
Reanalyses such as ERA5 have long been foundational for weather and climate science. They have also found a new use case, as training and verification data for machine-learnt weather prediction (MLWP) models. Here we compare short-lead time (6h) forecasts from the MLWP model GraphCast against ERA5. In doing so, we identify a...
Current techniques for post-training Large Language Models (LLMs) rely either on costly human supervision or on external verifiers to boost performance on tasks such as mathematical reasoning and code generation. However, as LLMs improve their problem-solving, any further improvement will potentially require high-quality solutions to...
Sequential-Horizon Vision-and-Language Navigation (SH-VLN) presents a challenging scenario where agents should sequentially execute multi-task navigation guided by complex, long-horizon language instructions. Current vision-and-language navigation models exhibit significant performance degradation with such multi-task instructions, as...
Travel planning is a sophisticated decision-making process that requires synthesizing multifaceted information to construct itineraries. However, existing travel planning approaches face several challenges: (1) Pruning candidate points of interest (POIs) while maintaining a high recall rate; (2) A single reasoning path restricts the...
Physical Unclonable Functions (PUFs) serve as lightweight, hardware-intrinsic entropy sources widely deployed in IoT security applications. However, delay-based PUFs are vulnerable to Machine Learning Attacks (MLAs), undermining their assumed unclonability. There are no valid metrics for evaluating PUF MLA resistance, but empirical...
In the process of digital transformation, enterprises are faced with problems such as insufficient semantic understanding of unstructured data and lack of intelligent decision-making basis in driving mechanisms. This study proposes a method that combines a large language model (LLM) and a knowledge graph. First, a fine-tuned BERT...
Out-of-distribution generalization in reinforcement learning is hard to diagnose when benchmark shifts mix dynamics, observations, goals, and rewards. We address this with Tape, a controlled benchmark that isolates latent rule-shift in dynamics while keeping the observation-action interface fixed. The protocol combines deterministic...
Knowledge distillation (KD) is one of the most effective paradigms for compressing large-scale foundation models into deployable architectures. In the context of Automatic Speech Recognition (ASR), previous studies have predominantly focused on forcing the student model to strictly mimic the predictive distribution of a massive...
Large Language Model-based Multi-Agent Systems (LLM-based MAS), where multiple LLM agents collaborate to solve complex tasks, have shown impressive performance in many areas. However, MAS are typically distributed across different devices or environments, making them vulnerable to perturbations such as agent failures. While existing...