arXiv preprints from January 1, 2026 through September 26, 2026 — 18:43:41 EST
Deep Reinforcement Learning (RL) agents often learn policies that achieve the same episodic return yet behave very differently, due to a combination of environmental (random transitions, initial conditions, reward noise) and algorithmic (minibatch selection, exploration noise) factors. In continuous control tasks, even small parameter...
To develop a reliable AI for psychological assessment, we introduce \texttt{PsychEval}, a multi-session, multi-therapy, and highly realistic benchmark designed to address three key challenges: \textbf{1) Can we train a highly realistic AI counselor?} Realistic counseling is a longitudinal task requiring sustained memory and dynamic...
Web applications (web apps) have become a key arena for large language models (LLMs) to demonstrate their code generation capabilities and commercial potential. However, building a benchmark for LLM-generated web apps remains challenging due to the need for real-world user requirements, generalizable evaluation metrics without relying...
The nature of the tetraneutron ($4n$) system remains a pivotal question in nuclear physics. We investigate the $4n$ system using nuclear lattice effective field theory in finite volumes with a lattice size up to $L=30$~fm, employing both a high-precision N$^3$LO interaction and a simplified SU(4) symmetric one. The ground-state energy...
Reinforcement learning (RL) has shown considerable potential in autonomous driving (AD), yet its vulnerability to perturbations remains a critical barrier to real-world deployment. As a primary countermeasure, adversarial training improves policy robustness by training the AD agent in the presence of an adversary that deliberately...
Diffusion policies have emerged as a powerful approach for robotic control, demonstrating superior expressiveness in modeling multimodal action distributions compared to conventional policy networks. However, their integration with online reinforcement learning remains challenging due to fundamental incompatibilities between diffusion...
This paper investigates a system of nonlinear reaction-diffusion equations modeling the industrial synthesis of ammonia. By applying Lie group analysis, we construct self-similar solutions and derive a reduced system of ordinary differential equations. Using comparison principles and barrier techniques, we establish sufficient...
Face verification systems have seen substantial advancements; however, they often lack transparency in their decision-making processes. In this paper, we introduce an innovative Vision-Language Model (VLM) for Face Verification, which not only accurately determines if two face images depict the same individual but also explicitly...
Given a probability space $(S,Δ, \mathbb{P})$ and a separable metric space $(U,d)$, the $Ky~Fan$ metric $ρ(X,Y)$ on the space $\mathfrak{X}^0$ of equivalence classes of random variables (w.r.t. almost sure equality) formed from the set $\mathfrak{X}(U)$ of $U$-valued random variables is given by $ρ(X,Y)=\inf...
Room-temperature, miniaturized, polarization-resolved terahertz (THz) detection of high speed is vital for high-resolution imaging in radar, remote sensing, and semiconductor inspection, and is essential for large-scale THz focal plane arrays. However, miniaturization below deep-subwavelength scales (< 1/50 wavelength) remain...
We present a new framework for analyzing the evolution of information in geophysical systems. Understanding how information, and its counterpart, uncertainty, propagates is central to predictability studies and has significant implications for applications such as forecast uncertainty quantification and risk management. It also offers...
We compute the gravitational fluxes and waveform for eccentric compact binaries including matter effects through adiabatic tidal interactions within the post-Newtonian approximation. The computations are performed at the relative 2.5PN order. Using the dynamics derived in the companion paper, we first derive the radiated energy and...
Federated learning is a privacy-focused approach towards machine learning where models are trained on client devices with locally available data and aggregated at a central server. However, the dependence on a single central server is challenging in the case of a large number of clients and even poses the risk of a single point of...
In this report, we present HyperCLOVA X 8B Omni, the first any-to-any omnimodal model in the HyperCLOVA X family that supports text, audio, and vision as both inputs and outputs. By consolidating multimodal understanding and generation into a single model rather than separate modality-specific pipelines, HyperCLOVA X 8B Omni serves as...
Semantic communication (SemCom) improves communication efficiency by transmitting task-relevant information instead of raw bits and is expected to be a key technology for 6G networks. Recent advances in generative AI (GenAI) further enhance SemCom by enabling robust semantic encoding and decoding under limited channel conditions....
The experimental advances in realizing artificial spin-orbit coupling (SOC) and non-Hermitian potentials in ultracold atomic system open a new avenue for exploring their significant roles in quantum many-body physics. Here, we investigate a non-Hermitian, two-component Fermi system in a cubic lattice with Rashba SOC and complex-valued...
This paper presents a general framework for end-to-end mutual information maximization in communication and sensing systems represented by stochastic directed acyclic graphs (DAGs). We derive a unified formula for the (mutual) information gradient with respect to arbitrary internal parameters, utilizing marginal and conditional score...
The centrosymmetric cubic compound EuTi$_2$Al$_{20}$, in which magnetic Eu ions form a diamond network, undergoes an antiferromagnetic transition at T$_N$ = 3.3 K and exhibits metamagnetic transitions at H$_{m1}$ = 1.7 T and H$_{m2}$ = 2.8 T for H || [100] at 1.9 K. Between these fields, the magnetization shows a step-like behavior,...
Lossy compression, widely used by scientists to reduce data from simulations, experiments, and observations, can distort features of interest even under bounded error. Such distortions may compromise downstream analyses and lead to incorrect scientific conclusions in applications such as combustion and cosmology. This paper presents a...
The ever-increasing adoption of Large Language Models in critical sectors like finance, healthcare, and government raises privacy concerns regarding the handling of sensitive Personally Identifiable Information (PII) during training. In response, regulations such as European Union's General Data Protection Regulation (GDPR) mandate...
Retrieval-Augmented Generation (RAG) systems often rely on fixed top-k document selection mechanisms that ignore downstream generation quality and impose computational overheads. We propose SRAS (Sparse Reward-Aware Selector), a lightweight document selector trained via reinforcement learning (RL) for edge-native RAG deployment....
The rapid evolution of AIGC technology enables misleading viewers by tampering mere small segments within a video, rendering video-level detection inaccurate and unpersuasive. Consequently, temporal forgery localization (TFL), which aims to precisely pinpoint tampered segments, becomes critical. However, existing methods are often...
The 2023 U.S. banking crisis propagated not through direct financial linkages but through a high-frequency, information-based contagion channel. This paper moves beyond exploration analysis to test the "too-similar-to-fail" hypothesis, arguing that risk spillovers were driven by perceived similarities in bank business models under...
How neural representations in the insular cortex support emotional processing remains poorly understood, and the extent to which the insula is specialized for disgust processing remains debated. We recorded stereoelectroencephalography data from the insula while human subjects with implanted electrode contacts performed a facial...
Self-supervised learning (SSL) methods have become a dominant paradigm for creating general purpose models whose capabilities can be transferred to downstream supervised learning tasks. However, most such methods rely on vast amounts of pretraining data. This work introduces Subimage Overlap Prediction, a novel self-supervised...