arXiv preprints from January 1, 2026 through September 16, 2026 — 17:26:57 EST
Fine-tuning LLMs on benign data can still degrade alignment and adversarial robustness, yet direct analysis of the role of fine-tuning objectives in shaping these safety outcomes remain limited. We present a controlled comparison of six fine-tuning objectives -- Supervised Fine-Tuning, Direct Preference Optimization, Conditional...
LIDAR 3D object detection is one of the important tasks for autonomous vehicles. Ensuring that this task operates in real-time is crucial. Toward this, model quantization can be used to accelerate the runtime. However, directly applying model quantization often leads to performance degradation due to LIDAR's wide numerical...
Many molecular properties depend on 3D geometry, where non-covalent interactions, stereochemical effects, and medium- to long-range forces are determined by spatial distances and angles that cannot be uniquely captured by a 2D bond graph. Yet most 3D molecular graph neural networks still rely on globally fixed neighborhood heuristics,...
Deploying Sentinel-2 satellite derived bathymetry (SDB) robustly across sites remains challenging. We analyze a Swin-Transformer based U-Net model (Swin-BathyUNet) to understand how it infers depth and when its predictions are trustworthy. A leave-one-band out study ranks spectral importance to the different bands consistent with...
Modeling the dynamics of open quantum systems on noisy intermediate-scale quantum (NISQ) devices constitutes a major challenge, as high noise levels and environmental degradations lead to the decay of pure quantum states (decoherence) and energy losses. This situation represents one of the most important problems in the field of...
Digital lending applications, commonly referred to as loan apps, have become a primary channel for microcredit in emerging markets. However, many of these apps demand excessive permissions and misuse sensitive user data for coercive debt-recovery practices, including harassment, blackmail, and public shaming that affect both borrowers...
Continual learning aims to enable neural networks to acquire new knowledge on sequential tasks. However, the key challenge in such settings is to learn new tasks without catastrophically forgetting previously learned tasks. We propose the Fisher-Orthogonal Projected Natural Gradient Descent (FOPNG) optimizer, which enforces...
Legal Judgment Prediction (LJP) aims to predict the outcomes of legal cases based on factual descriptions, serving as a fundamental task to advance the development of legal systems. Traditional methods often rely on statistical analyses or role-based simulations but face challenges with multiple allegations, diverse evidence, and lack...
We present the first unified framework for rate-distortion-optimized compression and segmentation of 3D Gaussian Splatting (3DGS). While 3DGS has proven effective for both real-time rendering and semantic scene understanding, prior works have largely treated these tasks independently, leaving their joint consideration unexplored....
FIRRTL is an intermediate representation language for Register Transfer Level (RTL) hardware designs. In FIRRTL programs, the bit widths of many components are not specified explicitly and must be inferred during compilation. In mainstream FIRRTL compilers, such as the official compiler firtool, width inference is conducted by a...
Clinical Question-Answering (CQA) industry systems are increasingly rely on Large Language Models (LLMs), yet their deployment is often guided by the assumption that domain-specific fine-tuning is essential. Although specialised medical LLMs such as BioBERT, BioGPT, and PubMedBERT remain popular, they face practical limitations...
The reproducibility of software environments is a critical concern in modern software engineering, with ramifications ranging from the effectiveness of collaboration workflows to software supply chain security and scientific reproducibility. Containerization technologies like Docker address this problem by encapsulating software...
We derive sharp bounds for the boundary control cost of the one-dimensional fractional Schrödinger and heat equations. The analysis of the lower bound is based on the study of the control cost of a related singular boundary control problem in finite time, using tools from complex analysis. The analysis of the upper bound relies on the...
Spatial understanding remains a key challenge in vision-language models. Yet it is still unclear whether such understanding is truly acquired, and if so, through what mechanisms. We present a controllable 1D image-text testbed to probe how left-right relational understanding emerges in Transformer-based vision and text encoders...
Conventional communication systems, including both separation-based coding and AI-driven joint source-channel coding (JSCC), are largely guided by Shannon's rate-distortion theory. However, relying on generic distortion metrics fails to capture complex human visual perception, often resulting in blurred or unrealistic reconstructions....
The emergent reasoning capabilities of Large Language Models (LLMs) offer a transformative paradigm for analyzing text-attributed graphs. While instruction tuning is the prevailing method for adapting pre-trained LLMs to graph learning tasks like node classification, it requires a substantial volume of annotated (INSTRUCTION, OUTPUT)...
We analyze the CMS-reported t tbar excess within conventional Two-Higgs-Doublet Models of Types I, II, X, and Y, using the best-fit pseudoscalar parameters MA = 365 GeV, GammaA over MA = 2 percent, and tan beta = 1.28. Applying theoretical and experimental constraints, including stability, unitarity, perturbativity, flavor...
Single-pass hallucination detectors rely on internal telemetry (e.g., uncertainty, hidden-state geometry, and attention) of large language models, implicitly assuming hallucinations leave separable traces in these signals. We study a white-box, model-side adversary that fine-tunes lightweight LoRA adapters on the model while keeping...
Large language models (LLMs) have shown growing promise in biomedical research, particularly for knowledge-driven interpretation tasks. However, their ability to reliably reason from gene-level knowledge to functional understanding, a core requirement for knowledge-enhanced cell atlas interpretation, remains largely underexplored. To...
Explainable AI (XAI) is crucial for building transparent and trustworthy machine learning systems, especially in high-stakes domains. Concept Bottleneck Models (CBMs) have emerged as a promising ante-hoc approach that provides interpretable, concept-level explanations by explicitly modeling human-understandable concepts. However,...
Recently, Zhang \cite{Zh20} and Li-Su-Wang-Zhang \cite{LSWZ24+} generalized Llarull's theorem to the noncompact complete spin manifold. In this paper, we further extend their results to the noncompact manifold with compact boundary.
We introduce UNMIXX, a novel framework for multiple singing voices separation (MSVS). While related to speech separation, MSVS faces unique challenges: data scarcity and the highly correlated nature of singing voices mixture. To address these issues, we propose UNMIXX with three key components: (1) musically informed mixing strategy...
Pigmentation abnormalities, ranging from hypo- to hyperpigmentation, can serve as biomarkers of developmental disruption in fish exposed to environmental contaminants. However, the mechanistic pathways underlying these alterations remain poorly understood. Studies have shown that pattern formation in fish development requires specific...
This paper presents an experimental study of microwave single-photon transitions that are magnetic-field-insensitive in degenerate Fermi gases of $^{40}$K. This contrasts with microwave single-photon clock transitions for 0-0 magnetic-field-insensitive states and two-photon clock transitions for non 0-0 magnetic-field-insensitive...
Humanoid robots are capable of performing various actions such as greeting, dancing and even backflipping. However, these motions are often hard-coded or specifically trained, which limits their versatility. In this work, we present FRoM-W1, an open-source framework designed to achieve general humanoid whole-body motion control using...