April 2026 arXiv papers — page 143

Seiya Mitsuno, Midori Ban, Hiroshi Ishiguro, Yuichiro Yoshikawa

Social isolation among older adults has become a critical concern, as reduced opportunities for conversation and weakened family relationships negatively affect mental health. This study proposes a dialogue agent that supports older adults by fostering both a relationship with the agent and a relationship with their grandchild through sharing everyday inform

Rong Wang, Ruyi Zha, Ziang Cheng, Jiayu Yang

We present a novel method for generating geometrically realistic and consistent orbital videos from a single image of an object. Existing video generation works mostly rely on pixel-wise attention to enforce view consistency across frames. However, such mechanism does not impose sufficient constraints for long-range extrapolation, e.g. rear-view synthesis, i

Haoran Li, Yulin Chen, Huihao Jing, Wenbin Hu

Individuals' concerns about data privacy and AI safety are highly contextualized and extend beyond sensitive patterns. Addressing these issues requires reasoning about the context to identify and mitigate potential risks. Though researchers have widely explored using large language models (LLMs) as evaluators for contextualized safety and privacy assessments

Ruiyang Xia, Qi Zhang, Yaowen Xu, Zhaofan Zou

The proliferation of highly realistic AI-Generated Image (AIGI) has necessitated the development of practical detection methods. While current AIGI detectors perform admirably on clean datasets, their detection performance frequently decreases when deployed "in the wild", where images are subjected to unpredictable, complex distortions. To resolve the critic

Utsho Kumar Dey

Pneumonia remains a leading cause of childhood mortality worldwide, with a heavy burden in low-resource settings such as Bangladesh where radiologist availability is limited. Most existing deep learning approaches treat pneumonia detection as a binary problem, overlooking the clinically critical distinction between bacterial and viral aetiology. This paper p

Koushik Chakraborty, Koyel Guha

This research paper addresses the limitations of semantic search in complex enterprise document ecosystems. Traditional RAG pipelines often fail to capture hierarchical and interconnected information, leading to retrieval inaccuracies. We propose Agentic Knowledge Graphs featuring Recursive Crawling as a robust solution for navigating superseding logic and m

Prasad Nimantha Madusanka Ukwatta Hewage, Hao Wu

Accurate short-term residential energy consumption forecasting at sub-hourly resolution is critical for smart grid management, demand response programmes, and renewable energy integration. While weather variables are widely acknowledged as key drivers of residential electricity demand, the relative merit of incorporating temporal autocorrelation - the sequen

Guofeng Cui, Yang Liu, Pichao Wang, Hankai Hsu

Batch active learning (BAL) is a crucial technique for reducing labeling costs and improving data efficiency in training large-scale deep learning models. Traditional BAL methods often rely on metrics like Mahalanobis Distance to balance uncertainty and diversity when selecting data for annotation. However, these methods predominantly focus on the distributi

Toshiaki Miyamoto

Gromov introduced the notion of a pyramid as a generalization of a metric measure space, based on the idea of the concentration of measure phenomenon. In this paper, we introduce the concept of a direct sum of pyramids, which naturally appears as a limit of a sequence of metric measure spaces whose measures concentrate on finitely or countably many regions,

Justice Owusu Agyemang, Jerry John Kponyo, Elliot Amponsah, Godfred Manu Addo Boakye

We present a systematic measurement study of seven tactics for reducing cloud LLM token usage when a small local model can act as a triage layer in front of a frontier cloud model. The tactics are: (1) local routing, (2) prompt compression, (3) semantic caching, (4) local drafting with cloud review, (5) minimal-diff edits, (6) structured intent extraction, a

Xi Wang, Tal Zussman, Yuang Xu, Bin Ma

Existing software-based memory tiering systems decide which pages to place on the slower or faster tier. However, they do not take into account two important factors that greatly influence application performance: the size of the migrated pages, and the underlying hardware device and tiering topology. We introduce TierBPF, a software mechanism that can be pl

Yu Jiang, Ching-Lung Lin, Gen Nakamura

Dynamic elastography is a widely used, safe, convenient, and cost-effective method to aid in medical diagnosis. It visualizes the wave field propagating through living tissues and quantitatively determines the wave propagation speed from the acquired data, thereby enabling the extraction of the viscoelastic properties of in vivo tissues. Notably, this identi

Yimin Lv, Shuli Wang, Beihong Jin, Yisong Yu

User behavior sequence modeling plays a significant role in Click-Through Rate (CTR) prediction on e-commerce platforms. Except for the interacted items, user behaviors contain rich interaction information, such as the behavior type, time, location, etc. However, so far, the information related to user behaviors has not yet been fully exploited. In the paper

Shiu Mochiyama, Rikuto Kawasome

The transition toward software-defined vehicles requires standardization and modularization of hardware decoupled from software, along with centralized electrical/electronic architectures. While electrified drive units, such as integrated in-wheel drives, are expected to realize the hardware standardization and unprecedented flexibility in vehicle design, th

Leonard Logarić, John Goold, Shane Dooley

One of the promising applications of digital quantum processors is the simulation of many-body quantum systems. They have been already used to investigate several ergodicity violating mechanisms, which were initially discovered in synthetic quantum matter, such as many-body localisation, Hilbert space fragmentation and quantum many-body scars (QMBS). In addi

Kaijie Yang, Huiyuan Zheng, Xiaodong Xu, Di Xiao

We show that excitons forming between moir\'e flat Chern bands possess a substantial electric dipole moment comparable to the moir\'e lattice parameter times the elementary charge ($\sim10^2$ Debye). At a hole filling factor of one in twisted MoTe$_2$, the dipole moment of the lowest-energy exciton branch develops in-plane helical texture in momentum space f

Kareem Soliman

Multi-site autism neuroimaging studies routinely control for the confound between full-scale IQ and head motion by regressing framewise displacement against IQ scores and removing shared variance. This procedure assumes that ordinary least squares (OLS) provides an unbiased estimate of the confound magnitude. We tested this assumption on the ABIDE-I phenotyp

Jose Gonzalez-Belmonte, Jaerock Kwon

As the number of fatalities involving Autonomous Vehicles increase, the need for a universal method of communicating between vehicles and other agents on the road has also increased. Over the past decade, numerous proposals of external Human-Machine Interfaces (eHMIs) have been brought forward with the purpose of bridging this communication gap, with none ye

Gaoxiang Cong, Liang Li, Jiaxin Ye, Zhedong Zhang

Movie dubbing aims to synthesize speech that preserves the vocal identity of a reference audio while synchronizing with the lip movements in a target video. Existing methods fail to achieve precise lip-sync and lack naturalness due to explicit alignment at the duration level. While implicit alignment solutions have emerged, they remain susceptible to interfe

Gautam Neelakantan Memana

We establish a comparison principle for viscosity subsolutions and supersolutions of a broad class of second-order quasilinear, maximally subelliptic PDEs on general manifolds. In fact, we prove the comparison theorem for a larger class of degenerate subelliptic PDEs. Our result strengthens a recent theorem of Manfredi-Mukherjee, which was established in the

Yizhe Chi, Deyao Hong, Dapeng Jiang, Tianwei Luo

Current LLM agent benchmarks, which predominantly focus on binary pass/fail tasks such as code generation or search-based question answering, often neglect the value of real-world engineering that is often captured through the iterative optimization of feasible designs. To this end, we introduce Frontier-Eng, a human-verified benchmark for generative optimiz

Manuel Tonneau, Dylan Thurgood, Diyi Liu, Niyati Malhotra

Online hate speech is associated with harms ranging from deteriorating mental health to violence, yet how consistently platforms moderate hate, and whether enforcement is feasible at scale, remain poorly understood. We audit hate speech moderation on Twitter (now X) using 540,000 tweets annotated by trained native speakers, representative of a full day on th

Chris Albert, Ritoban Basu Thakur, Farzad Faramarzi, Byeong Ho Eom

We present a method of spatially mapping microwave kinetic inductance detector (MKID) arrays, in a dark setup. MKIDs are superconducting natively multiplexed resonators which enable kilopixel arrays, such as for the proposed Probe far-Infrared Mission for Astrophysics (PRIMA). In such telescope applications one must map the spatial location of each MKID with

Yutao Ma, Xujia Meng

We consider the product of \(k_{n}\) independent \(n\times n\) complex Ginibre matrices and denote its eigenvalues by \(Z_{1},\ldots ,Z_{n}\). Let \(\alpha = \lim_{n\to\infty} n / k_{n}\). Using the determinantal point process method, we reduce the study of extremal eigenvalues to the evaluation of determinants of certain \(n\times n\) matrices. In the modul

Clara Xue, Zizheng Yan, Zhenning Shi, Yuhang Yu

Live Photo captures both a high-quality key photo and a short video clip to preserve the precious dynamics around the captured moment. While users may choose alternative frames as the key photo to capture better expressions or timing, these frames often exhibit noticeable quality degradation, as the photo capture ISP pipeline delivers significantly higher im

Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu

To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise. Conversely, discrete structured memory architectures provide

Yulin Chen, Tri Cao, Haoran Li, Yue Liu

Web agents powered by vision-language models (VLMs) enable autonomous interaction with web environments by perceiving and acting on both visual and textual webpage content to accomplish user-specified tasks. However, they are highly vulnerable to prompt injection attacks, where adversarial instructions embedded in HTML or rendered screenshots can manipulate

Vangara Saiprudhvi, Sandeep Singh, Keshav Singh, Hariharan Subramaniyam

Integrated terrestrial and non-terrestrial networks (ITNTNs) are regarded as a key architectural paradigm for sixth-generation (6G) wireless systems. This paper investigates a dual-aerial reconfigurable intelligent surface (RIS)-assisted ITNTN, where a terrestrial base station (TBS) and a satellite (SAT) jointly serve terrestrial and satellite users with the

Houxing Ren, Mingjie Zhan, Zimu Lu, Ke Wang

Spreadsheets are central to real-world applications such as enterprise reporting, auditing, and scientific data management. Despite their ubiquity, existing large language model based approaches typically treat tables as plain text, overlooking critical layout cues and visual semantics. Moreover, real-world spreadsheets are often massive in scale, exceeding

Dongkyung Kang, Jaeyeon Hwang, Junseo Park, Minji Kang

Style transfer applies the appearance of a reference image to a content image while preserving its spatial structure. Recent diffusion-based methods achieve strong stylization but typically assume a single global style. We instead consider regional-multi style transfer, which assigns multiple references to user-specified regions of a content image. Extending

Yutaro Mikami

In the previous paper of the author, motivated by the categorical $p$-adic local Langlands correspondence, the author studied families of $G_K$-equivariant vector bundles over the Fargues-Fontaine curve parametrized by algebraic-affinoid $\mathbb{Q}_{p,\square}$-algebras (e.g., $\mathbb{Q}_p[T]$). In this paper, we study the $p$-adic Hodge theoretic properti

K. V. Kozenko, V. V. Gromyko, I. I. Beterov, I. I. Ryabtsev

We numerically optimized a scheme for a neutral atom Rydberg blockade symmetric controlled-Z (CZ) gate to increase its robustness to variations in the Rabi frequency. This gate scheme uses analytically defined phase profiles of the laser pulse and demonstrates increased robustness to variations in the Rabi frequency almost by an order of magnitude compared t

Hailong Gong, Haibo Zhang, Amanda S. Barnard, Mahbub Hassan

Matrix multiplication is a fundamental kernel in large-scale artificial intelligence and scientific computing, but its performance on conventional electronic accelerators is increasingly constrained by memory bandwidth and energy efficiency. Photonic computing offers a promising alternative due to its ultra-high bandwidth, massive parallelism, and low power

Satyam Kumar, Saurabh Jha

We present Forge-UGC (FX Optimization and Register-Graph Engine for Universal Graph Compilation), a four-phase compiler for transformer deployment on heterogeneous accelerator hardware, validated on Intel AI Boost NPU. Existing frameworks such as OpenVINO and ONNX Runtime often use opaque compilation pipelines, limited pass-level visibility, and weak buffer

Firdous Haidar, Muhammed Jaffar A., Ipsita Mandal

We compute the magnetoelectric conductivity of a semimetal hosting an ideal gapped nodal ring (GNR) in three distinct planar-Hall configurations, in the simultaneous presence of an external electric field $\boldsymbol{E}$, a magnetic field $\boldsymbol{B}$, and a strain-induced axial pseudomagnetic field $\boldsymbol{B}_5$. The latter arises from a nonunifor

Fumihiko Asano, Ning Lei, Taiki Sedoguchi

A class of planar bipedal robots with unique mechanical properties has been proposed, where all links are balanced around the hip joint, preventing natural swinging motion due to gravity. A common property of their equations of motion is that the inertia matrix is a constant matrix, there are no nonlinear velocity terms, and the gravity term contains simple

Yexiong Lin, Jia Shi, Shanshan Ye, Wanyu Wang

Flow matching has emerged as a powerful generative framework, with recent few-step methods achieving remarkable inference acceleration. However, we identify a critical yet overlooked limitation: these models suffer from severe diversity degradation, concentrating samples on dominant modes while neglecting rare but valid variations of the target distribution.

Ayan Kumar Nai, G. K. Samanta

Entanglement-based quantum key distribution (QKD) relies on the distribution of high-fidelity maximally entangled Bell states, typically generated via spontaneous parametric down-conversion (SPDC). In practical systems, unwanted relative phases arise from birefringence, pump-beam contributions, imperfect photon-pair generation, transmission through physical

Yilong Zuo, Xunkai Li, Zhihan Zhang, Ronghua Li

Multimodal attributed graphs (MAGs) combine multimodal node attributes with structured relations. However, existing methods usually perform shared message passing on a single graph and implicitly assume that the same neighbors are equally useful for all modalities. In practice, neighbors that benefit one modality may interfere with another, blurring modality

Yuan Huang, Sijie Zhao, Jing Cheng, Hao Xu

Stereo video inpainting, which aims to fill the occluded regions of warped videos with visually coherent content while maintaining temporal consistency, remains a challenging open problem. The regions to be filled are scattered along object boundaries and occupy only a small fraction of each frame, leading to two key challenges. First, existing approaches pe

Sergey Dyachenko, Robert Marangell, Dmitry E. Pelinovsky

We study the modulational stability problem for the traveling periodic waves (called Stokes waves) in an infinitely deep fluid by using pseudo-differential operators in conformal variables. We derive the criteria and the normal forms for four bifurcations which are repeated recurrently when the steepness of the Stokes wave is increased towards the highest wa

Zaoyu Chen, Jianbo Dai, Boyu Zhu, Jingdong Wang

Large language models (LLMs) can generate code from natural language, but the extent to which they capture intended program behavior remains unclear. Executable behavioral specifications, defined via preconditions and postconditions, provide a concrete means to assess such understanding. However, existing work on specification generation is constrained in ev

Arul Lakshminarayan, Karol Życzkowski

The notion of Shannon entropy is crucial for the theory of classical information. In quantum information theory, an analogous key role is played by the von Neumann entropy: quantum information processing is closely related to entropy dynamics. This reveals a direct link with the theory of quantum chaotic systems, which can be characterized by a positive entr

Vangara Saiprudhvi, Keshav Singh, Hariharan Subramaniyam, Chih-Peng Li

Integrated terrestrial and non-terrestrial networks (ITNTNs) are envisioned as a key paradigm for sixth-generation (6G) wireless systems, enabling seamless global connectivity. In this paper, we investigate a dual-aerial active reconfigurable intelligent surface (ARIS)-assisted non-orthogonal multiple access (NOMA)-based ITNTN, where a terrestrial base stati

Malik Amir

We present an expanded expository account of the $K$-moment problem for polynomial algebras over \(\R^d\), with special emphasis on compact basic closed semialgebraic sets. The central question is to characterize those linear functionals on \(\R[x_1,\dots,x_d]\) which admit representation by integration against a positive Radon measure supported on a prescri

Ian Babich, Timofey M. Savilov, Natalia A. Mamchik, Kristina Vaklinova

The advent of van der Waals (vdW) heterostructures has enabled formation of bespoke materials with atomic precision, where numerous quantum and topological phenomena have already been discovered. This atomic-layer tunability, however, comes at a cost: individual 2D layers must be picked up, moved, and placed in a deterministic manner while keeping their inte

Jiyuan Tan, Jose Blanchet, Vasilis Syrgkanis

Policy-Relevant Treatment Effects (PRTEs) are generally not point-identified under standard Instrumental Variable (IV) assumptions when the instrument generates limited support in treatment propensity. We show that PRTE partial identification in the generalized Roy model can instead be formulated as a Constrained Conditional Optimal Transport (CCOT) problem

Raeyoung Chang, Dongwook Kwon, Jisoo Lee, Nikhil Verma

Cascaded LLM systems coordinate models of varying sizes with human experts to balance accuracy, cost, and abstention under uncertainty. However, single-model tiers at each stage often struggle with ambiguous queries, triggering premature escalations to costlier models or experts due to under-confidence and inefficient compute scaling. CascadeDebate addresses

Peng Zhao, Peng Xu, Zheng-Yuan Xue

The path toward practical superconducting quantum processors requires the integration of a large number of high-performance qubits. Modular architectures could offer a way to address the scaling limitations of monolithic designs by partitioning a large quantum processor into physically separated modules, or chiplets, linked through long-range interconnects.

Zonghong Liu, Matthew Dwyer, Salim El Rouayheb

We study decentralized learning over networks where data are distributed across nodes without a central coordinator. Random walk learning is a token-based approach in which a single model is propagated across the network and updated at each visited node using local data, thereby incurring low communication and computational overheads. In weighted random-walk

Mohsen Amini Salehi, Adel N. Tousi, Hai Duc Nguyen, Murtaza Rangwala

Advances in networking and computing technologies throughout the early decades of the 21st century have transformed long-standing dreams of pervasive communication and computation into reality. These technologies now form a rapidly evolving and increasingly complex global infrastructure that will underpin the next aspiration of computing: supporting intellig

Taehun Kim, Hyeryun Park, Hyeonhoon Lee, Yushin Lee

Clinical data-driven research requires clinical expertise, programming skills, access to patient data, and extensive documentation, creating barriers and slowing the pace for clinicians and external researchers. To address this, we developed the Clinical Agentic Research Intelligence System (CARIS) that automates the workflow: research planning, literature s

Jialu Liu, Yao Li, Zhuoheng Li, Huining Li

Augmented reality (AR) systems pose unique privacy risks due to their continuous capture of visual data. Existing AR privacy frameworks lack semantic understanding of visual content, limiting their effectiveness in detecting context-dependent privacy risks. We propose PrivAR, which leverages vision language models (VLMs) with chain-of-thought prompting for c

Mohammad AlShaikh Saleh, Sanjay Chawla, Sertac Bayhan, Haitham Abu-Rub

AI data centers experience rapid fluctuations in power demand due to the heterogeneity of computational tasks that they have to support. For example, the power profile of inference and training of large language models (LLMs) is quite distinct and big divergences can result in the instability of the underlying electricity grid. In this paper we propose, to t

Hang Xu, Chen Long, Bing Wang, Hao Chen

Underwater Image Enhancement (UIE) is essential for robust visual perception in marine applications. However, existing methods predominantly rely on uniform mapping tailored to average dataset distributions, leading to over-processing mildly degraded images or insufficient recovery for severe ones. To address this challenge, we propose a novel adaptive enhan

Chuan-Chi Wang, Yan-Jie Wang, Chia-Heng Tu, Shih-Hao Hung

Quantum circuit simulation is crucial for the development of quantum algorithms, particularly given the high cost and noise limitations of physical quantum hardware. While full-state quantum circuit simulation is commonly employed for prototyping and debugging, it poses challenges because of the exponential increase in simulation time for large quantum syste

Ali Mollahosseini, Mohammed Haroon Dupty, Wee Sun Lee

Accurate prediction of energy and forces for 3D molecular systems is one of fundamental challenges at the core of AI for Science applications. Many powerful and data-efficient neural networks predict molecular energies and forces from single atomic configurations. However, one crucial aspect of the data generation process is rarely considered while learning

Ping Zhao, Hongfei Wang

This paper develops a new framework for alpha testing in high-dimensional factor pricing models with time-varying coefficients. To detect sparse alternatives, we propose a spatial-sign-based max-type test and derive its limiting null distribution. A key theoretical result is that our statistic is asymptotically independent of the spatial-sign-based sum-type

Xinliang Wang, Yifeng Shi, Zhenyu Wu

3D Gaussian Splatting (3DGS) delivers high-fidelity real-time rendering but suffers from geometric and photometric degradations under sparse-view constraints. Current generative restoration approaches are often limited by insufficient temporal coherence, a lack of explicit spatial constraints, and a lack of large-scale training data, resulting in multi-view

Haiyan Lu

We systematically investigate the electronic structures of americium (Am), curium (Cm), berkelium (Bk), and californium (Cf) in both the ambient-pressure double hexagonal close-packed (dhcp) and high-pressure face-centered cubic (fcc) phases, using density functional theory combined with embedded dynamical mean-field approach. Our results reveal that Am exhi

Jiaqi Fan, Fan Yang, Jun Yin

We study large $N\times N$ power-law random band matrices $H=(H_{ij})$ with centered complex Gaussian entries, where the variances satisfy a power-law decay $\mathbb{E}|H_{ij}|^2\propto (|i-j|/W+1)^{-1-α}$, for some exponent $α>-1$ and bandwidth $W\gg 1$. We establish the following lower bounds, with high probability, on the localization length $\ell$ of bul

Zhuofan Wen, Yang Feng

Speculative decoding has emerged as a promising approach to accelerate autoregressive inference in large language models (LLMs). Self-draft methods, which leverage the base LLM itself for speculation, avoid the overhead of auxiliary draft models but face limitations: shallow layers often produce overconfident yet incorrect token predictions, and the presence

Hsin-Tien Chiang, John H. L. Hansen

Speech enhancement (SE) is critical for improving speech intelligibility and quality in real-world environments, particularly for cochlear implant (CI) users who experience severe degradations in speech understanding under noisy and reverberant conditions. In this study, we propose TokenSE, a discrete token-based SE framework operating in the neural audio co

Sandra Gómez-Gálvez, Tobias Olenyi, Gillian Dobbie, Katerina Taškova

Deep neural networks, despite their high accuracy, often exhibit poor confidence calibration, limiting their reliability in high-stakes applications. Current ad-hoc confidence calibration methods attempt to fix this during training but face a fundamental trade-off: two-phase training methods achieve strong classification performance at the cost of training i

Nima Alibabaei

We consider a finite family of invertible $2 \times 2$ real matrices and a transitive Markov shift on the index set. Let $\lambda$ be the top Lyapunov exponent for random matrix products driven by the Markov shift. We prove that, if the matrices are projectively uniformly hyperbolic with respect to the Markov shift, then $\lambda$ admits an explicit represen

María José Colmenares, Edwin A. Bergin, Ke Zhang, Geoffrey A. Blake

We present JWST/MIRI-MRS observations of ISO-Oph 37, a highly inclined flat-spectrum ($\lesssim$1 Myr old) source, to investigate the chemical composition and dynamical origin of its inner-disk gas. The spectrum reveals a rich combination of molecular emission and absorption: H$_2$O, CO, and OH are detected in emission, while strong absorption is observed fr

Ehsan Kamali Ahangar, Dominic Robe, Elnaz Hajizadeh

Amphiphilic block copolymers self-assemble into diverse nanoscale morphologies with significant implications for drug delivery. This work presents systematic Brownian dynamics simulations of multi-chain diblock and triblock copolymers across dilute and semi-dilute unentangled regimes, hydrophobic fractions, f of 0-1, and shear rates of 0-0.1 1/ns. In the dil

Haojie Ye, Arjun Laxman, Yichao Yuan, Krisztian Flautner

Money laundering detection faces challenges due to excessive false positives and inadequate adaptation to sophisticated multi-stage schemes that exploit modern financial networks. Graph analytics and AI are promising tools, but they struggle with the fuzziness of laundering patterns, which exhibit structural and temporal variations. Conventional data mining

Yigang Zheng

This paper investigates the Brennan Conjecture for domains $\Omega$ that arise as basins of attraction of a polynomial. We extend the result of Baranski, Volberg, and Zdunik to a broader class of polynomials. We prove that for any monic polynomial of degree $m$ with sufficiently small non-leading coefficients, Brennan Conjecture holds for its basin of attrac

Vincent Wesley Couey

Many engineering failures (thermal hotspot concentration, Hertz contact fatigue localization, boundary-layer loss, mixing dead zones) are geometric failure modes: changing the material delays the failure; changing the geometry eliminates it. Despite this, no formal metric exists for evaluating how uniformly a convex body distributes surface contact during ro

Ziqing Wang, Yibo Wen, Abhishek Pandy, Han Liu

In drug discovery, molecular optimization aims to iteratively refine a lead compound to improve molecular properties while preserving structural similarity to the original molecule. However, each oracle evaluation is expensive, making sample efficiency a key challenge for existing methods under a limited oracle budget. Trial-and-error approaches require many

Yang Cai, Weiqiang Zheng

We study the monotone inclusion problem $0\in F(z)+A(z)$, where $F$ is monotone and Lipschitz, and $A$ is maximally monotone, a framework that encompasses monotone variational inequalities and convex-concave saddle-point problems with constraints or regularization. It is well known that vanilla gradient descent diverges for this problem, whereas optimism-bas

Samrendra Roy, Kazuma Kobayashi, Souvik Chakraborty, Sajedul Talukder

Continual learning, the ability to acquire new tasks sequentially without forgetting prior knowledge, is essential for deploying neural networks in dynamic real-world environments, from nuclear digital twin monitoring to grid-edge fault detection. Existing synaptic importance methods, such as Elastic Weight Consolidation (EWC) and Synaptic Intelligence (SI),

Ziliang Wang, Gaoyun Lin, Xuesi Wang, Shaoqiang Liang

Generative Recommendation (GR) reframes retrieval and ranking as autoregressive decoding over Semantic IDs (SIDs), unifying the multi-stage pipeline into a single model. Yet a fundamental expressive gap persists: discriminative models score items with direct feature access enabling explicit user-item crossing, whereas GR decodes over compact SID tokens witho

Dongbin Li, Alexander E. Litvak, Tingzhou Yu

Let $M$ be an $n\times n$ random matrix with entries in $\{0, 1\}$, where each row is independently and uniformly sampled from the set of all vectors in $\{0, 1\}^n$ containing exactly $d$ ones, with $d=pn$ for some fixed constant $p\in (0,1/2]$. A recent result of Tran states that the smallest singular value $s_n(M)$ is bounded below by $c_p n^{-1/2}$ with

Qingchao Shen, Zibo Xiao, Lili Huang, Enwei Hu

Large Language Models (LLMs) are increasingly deployed across diverse domains, yet their vulnerability to jailbreak attacks, where adversarial inputs bypass safety mechanisms to elicit harmful outputs, poses significant security risks. While prior work has primarily focused on prompt injection attacks, these approaches often require resource-intensive prompt

Tao Feng, Pengrui Han, Guanyu Lin, Ge Liu

Large language models (LLMs) have transformed AI research thanks to their powerful internal capabilities and knowledge. However, existing LLMs still fail to effectively incorporate the massive external knowledge when interacting with the world. Although retrieval-augmented LLMs are proposed to mitigate the issue, they are still fundamentally constrained by t

Agnid Banerjee, Nicola Garofalo

In this note we derive a space-like quantitative uniqueness result for parabolic operators with H\"older zero-order term that interpolates between the Donnelly-Fefferman and the Bourgain-Kenig estimate. This generalizes a recent result of Teng, Wang and Zhu for the time-independent Schr\"odinger operator with a H\"older potential.

Jawad Hossain, Xiangyu Guo, Jiawei Zhou, Chong Liu

Small language models (SLMs) often struggle with complex mathematical reasoning due to limited capacity to maintain long chains of intermediate steps and to recover from early errors. We address this challenge by introducing a hint-assisted reasoning framework that incrementally guides SLMs through multi-step mathematical problem solving. Our approach decomp

Pei-Yu Tseng, Lan Zhang, ZihDwo Yeh, Xiaoyan Sun

Cyber Threat Intelligence (CTI) reports contain Indicators of Compromise (IOCs) that are critical for security operations. To operationalize these IOCs across heterogeneous logs, analysts often convert them into regular expressions (regexes) for tasks such as digital forensics, log parsing, and SIEM rule creation. However, regex construction is still largely

Xiuxiu Tang, G. Alex Ambrose, Ying Cheng

Student responses in STEM assessments are often handwritten and combine symbolic expressions, calculations, and diagrams, creating substantial variation in format and interpretation. Despite their importance for evaluating students' reasoning, such responses are time-consuming to score and prone to rater inconsistency, particularly when partial credit is req

Abey López-García, Erwin Miña-Díaz

For the Riesz and logarithmic energies, we consider a greedy sequence $(a_n)_{n=0}^\infty$ of points on the unit circle $S^1$ constructed in such a way that for every integer $N\geq 2$, the energy of the configuration $(a_0,\ldots,a_{N-2},x)$ attains its optimal value (say $E_N$) at $x=a_{N-1}$. We derive an asymptotic expansion for $E_N$ in terms of certain

Anand Aruna Kumar

This work investigates the dynamics of a charged particle in a uniform magnetic field within the Bohm--Madelung formulation of quantum mechanics. In this representation, the stationary Schrodinger equation separates into coupled amplitude and phase equations, where the amplitude sector admits a Sturm--Liouville structure supporting Ermakov--Lewis invariants.

Jiechao Gao, Rohan Kumar Yadav, Yuangang Li, Yuandong Pan

Pretrained language models (PLMs) like BERT provide strong semantic representations but are costly and opaque, while symbolic models such as the Tsetlin Machine (TM) offer transparency but lack semantic generalization. We propose a semantic bootstrapping framework that transfers LLM knowledge into symbolic form, combining interpretability with semantic capac

Jun Takeshita, Kei Hirose, Suyu Fu, Fumiya Sakai

It has been assumed that hydrogen is negligibly incorporated into core-forming metals below $\sim$3 GPa, and therefore the presence of hydrogen in iron cores of small terrestrial bodies including the moon has not been considered. Here we performed high-pressure melting experiments on the Fe-H system under H$_2$-saturated conditions, combined with synchrotron

Qingyuan Cai, Saihui Hou, Xuecai Hu, Yongzhen Huang

Gait recognition, as a reliable biometric technology, has seen rapid development in recent years while it faces significant challenges caused by diverse clothing styles in the real world. This paper introduces BarbieGait, a synthetic gait dataset where real-world subjects are uniquely mapped into a virtual engine to simulate extensive clothing changes while

Chenyan Liu, Yun Lin, Yuhuan Huang, Jiaxin Chang

In industrial and open-source software engineering tasks, developers often perform project-wise code editing tasks, including feature enhancement, refactoring, and bug fixing, where the leading AI models are expected to support the productivity. Hence, researchers and practitioners have proposed and adopted many LLM-based solutions to facilitate their real-w

Wentai Zhang, Ronghui Xi, Shiyao Peng, Jiayu Huang

Video Diffusion Transformers have revolutionized high-fidelity video generation but suffer from the massive computational burden of self-attention. While sparse attention provides a promising acceleration solution, existing methods frequently provoke severe visual flickering caused by static sparsity patterns and deterministic block routing. To resolve these

Disha Patel

System log anomaly detection is critical for maintaining the reliability of large-scale software systems, yet traditional methods struggle with the heterogeneous and evolving nature of modern log data. Recent advances in Large Language Models (LLMs) offer promising new approaches to log understanding, but a systematic comparison of LLM-based methods against

Yuxuan Wang, Qibiao Li, Youcheng Cai

Volumetric visualization has long been dominated by Direct Volume Rendering (DVR), which operates on dense voxel grids and suffers from limited scalability as resolution and interactivity demands increase. Recent advances in 3D Gaussian Splatting (3DGS) offer a representation-centric alternative; however, existing volumetric extensions still depend on costly

Shangkun Che, Silin Du, Ge Gao

The widespread use of Large Language Models (LLMs) in text generation has raised increasing concerns about intellectual property disputes. Watermarking techniques, which embed meta information into AI-generated content (AIGC), have the potential to serve as judicial evidence. However, existing methods rely on statistical signals in token distributions, leadi

Xianping Li, Woinshet D. Mergia, Kailash C. Patidar

In ecological studies of pattern formation, models of the competitive-diffusion type are generally singularly perturbed, and the numerical approximation of such models is challenging. In this paper, we present finite element discretization combined with a second-order semi-implicit nonlinear multistep scheme for a two-dimensional three-species competition-di

Yang Liu, Da Song, Armstrong Foundjem, Heng Li

Chain-of-Thought (CoT) prompting is widely used to elicit explicit reasoning from large language models for code (LLM4Code). However, its impact on robustness and the stability of reasoning trajectories under realistic input perturbations remains poorly understood. Prior work has largely evaluated CoT through final correctness, leaving a critical gap in unde

Vasundra Srinivasan

Preserving multimodal signals across agent boundaries is necessary for accurate cross-modal reasoning, but it is not sufficient. We show that modality-native routing in Agent-to-Agent (A2A) networks improves task accuracy by 20 percentage points over text-bottleneck baselines, but only when the downstream reasoning agent can exploit the richer context that n

Weikang Zhang, Zimo Zhu, Zhichuan Yang, Chen Huang

Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation o

Donglin Zhu, Peiyao Li, Ge Jin

The increasing demand for deep learning in seismic interpretation has highlighted significant challenges, particularly the reliance on massive, labeled datasets and the inefficiency of training isolated models for individual tasks. To address these limitations, we introduce a unified, prompt-guided flow-matching framework (SeisDiff-intp) capable of executing

Zhihua Hua, Junli Wang, Pengfei LI, Qihao Jin

Global navigation information and local scene understanding are two crucial components of autonomous driving systems. However, our experimental results indicate that many end-to-end autonomous driving systems tend to over-rely on local scene understanding while failing to utilize global navigation information. These systems exhibit weak correlation between t

Stephanie Kwari Dharmaputri, Anish Nagpal, Greg Nyilasy, Jing Lei

Advancements in Artificial Intelligence (AI) technologies' social fluency are being integrated into commercial interactions. As tools such as OpenAI's assistant are integrated into platforms such as Shopify, Klarna, and Visa, understanding consumer responses to AI social features become essential. One such feature is relational talk, an informal and non-obli

Tokuji Araya, Naoya Hiramatsu, Ryo Takahashi

We study reducing invariants of modules related to certain homological properties. For modules of finite reducing projective dimension, we establish grade inequalities. We prove that if $\mathbb{P}$ is the (uniform) Auslander condition, or the generalized Auslander--Reiten conjecture, or dependence of the total reflexivity conditions, then a module satisfies

Xue Zhao, Xiao Wang, Wentao Yang, Xiaoyu Dai

We describe a simple and compact experimental setup for optical tweezer arrays of 87Rb atoms. This setup includes a compact vacuum system, a single cooling laser, a simple tweezer laser, and a flexible control system. The small vacuum system with only 40 cm length takes advantage of the high atomic flux two-dimensional magneto-optical trap (2D MOT) while mai