arXiv subjects

Xinyi Zhao

Publications and source records attributed to Xinyi Zhao.

At least 19 recordsLinked to original sources

arXiv2026

OTTER - Two Transistor - One RRAM Architecture for Reliable In-Memory-Computing in 28 nm CMOS Technology

This work presents OTTER, a 28 nm CMOS platform co-integrated with TaOx-based valence-change mechanism (VCM) RRAM, demonstrating a two-transistor-one-memristive-device (2T1R) architecture for reliable in-memory computing. The 2T1R cell combines a low-drive-current (LD) transistor and a high-drive-current (HD) transistor in parallel, providing dedicated bias paths for SET programming and RESET operation, respectively. Through systematic experimental and simulated comparison of various transistor-pairing configurations using the physical compact model JART VCM Rth, design guidelines for transistor sizing are derived, establishing the minimum RESET transistor W/L required for complete RESET as a function of the SET current compliance. The 2T1R cell is further characterized under pulse-based programming, demonstrating multilevel analog conductance tuning with narrow, well separated conductance states across six programmable levels. An analog content-addressable memory (aCAM) design based on the same 2T1R cell is additionally analyzed at the circuit level, evaluating trade-offs between top- and bottom-connected RRAM comparator configurations. A hardware implementation of compute-in-memory (CIM) multiply-and-accumulate (MAC) operations is further demonstrated on a 15 x 15 2T1R crossbar array.

cs.ET↗

arXiv2026

Open-Source Autonomous Driving System Analysis and Multi-Disciplinary Hardware-in-the-Loop Research Paradigm with Reinforcement-Learning Testing and Large Language Models

Open-source autonomous driving systems provide an inspectable software foundation for intelligent vehicle research. Under real-vehicle deployment conditions, the recording and review of experimental conditions are important for interpreting system behavior and reusing experimental results. However, in a shared real-vehicle environment involving multiple vehicles, task processes, code modifications, and hardware testing feedback are often distributed across different teams and experimental stages, making it challenging to maintain continuous and reviewable experimental records. To address this limitation, this paper examines an Apollo-on-Hongqi EV environment and proposes a real-vehicle experimental framework. The framework connects multi-vehicle experiments, repository-based code reuse and software-hardware testing feedback within a unified review process. Large language models and RL-based testing serve as auxiliary components for record organization, anomaly summarization, and simulation-based candidate scenario generation. Based on this setting, this paper analyzes preliminary evidence from multi-vehicle collaborative experimentation, code and experimental-skill sharing, and software-hardware collaborative testing. The analysis shows that experimental records can be examined together with their operating conditions, providing a reviewable basis for Apollo-on-Hongqi EV research.

cs.SE↗

arXiv2026

DUET -- Dual User Embedding Transformers for Offsite Conversion Prediction

Offsite conversion rate (OCVR) prediction is an important ranking problem in computational recommendation systems. This task presents a modeling challenge: click signals are abundant and exhibit short temporal horizons, whereas conversion signals are inherently sparse, long-delayed, and frequently unattributed. Despite these statistical disparities, both signal types must inform models that operate within strict serving-latency constraints. Prior pre-training approaches address this heterogeneity with a single, undifferentiated encoder applied uniformly across both data streams. We propose DUET (Dual User Embedding Transformers), a framework that explicitly partitions user behavioral data into two domain-coherent streams -- clicks and conversions -- and pre-trains dedicated transformer encoders with architectures tailored to each stream's statistical characteristics: multi-layer self-attention for the dense click stream and interleaved cross- and self-attention for the sparse conversion stream. The resulting complementary embeddings are jointly consumed by a downstream ranker without exceeding serving-latency budgets. Evaluation demonstrates up to 0.38% normalized entropy (NE) reduction relative to the strongest baseline, and A/B test shows consistent improvements in OCVR prediction accuracy.

cs.LG↗

arXiv2026

A Markov Decision Process Framework for Enhancing Power System Resilience during Wildfires under Decision-Dependent Uncertainty

Wildfires pose an increasing threat to the safety and reliability of power systems, particularly in distribution networks located in fire-prone regions. To mitigate ignition risk from electrical infrastructure, utilities often employ safety power shutoffs, which proactively de-energize high-risk lines during hazardous weather and restore them once conditions improve. While this strategy can result in temporary load loss, it helps prevent equipment damage and wildfire ignition development in the system. In this paper, we develop a state-based decision-making framework to optimize such switching actions over time, with the goal of minimizing total operational costs throughout a wildfire event. The model represents network topologies as Markov states, with transitions influenced by both exogenous weather conditions and endogenous power flow dynamics. To address the computational challenges posed by the large state and action spaces, we propose an approximate dynamic programming algorithm based on post-decision states. The effectiveness and scalability of the proposed approach are demonstrated through case studies on 54-bus and 138-bus distribution systems, showcasing its potential for enhancing wildfire resilience across different grid configurations.

eess.SY↗

arXiv2026

An efficient compact splitting Fourier spectral methods for computing the dynamics of rotating spin-orbit coupled spin-2 Bose-Einstein condenstates

This paper investigates the dynamics of spin-2 Bose-Einstein condensates (BECs) with rotation and spin-orbit coupling (SOC). In order to better simulate the dynamics, we present an efficient high-order compact splitting Fourier spectral method. This method splits the Hamiltonian into a linear part, which consists of the Laplace, rotation and SOC terms, and a nonlinear part that includes all the remaining terms. The wave function is well approximated by the Fourier spectral method and is numerically accessed with discrete Fast Fourier transform (FFT). For linear subproblem, the handling of rotation term and SOC term poses a major challenge. Using a function mapping based on rotation, we can integrate the linear subproblem exactly and explicitly. This mapping we propose not only helps eliminate the rotation term, but also prevents the SOC term from evolving into a time-dependent form. The nonlinear subproblem is integrated analytically in physical space. Such "compact" splitting involves only two operators and facilitates the design of high-order splitting schemes. Our method is spectrally accurate in space and high order in time. It is efficient, explicit, unconditionally stable and simple to implement. In addition, we derive some dynamical properties and carry out a systematic study, including accuracy and efficiency tests, dynamical property verification, the SOC effects and dynamics of vortex lattice.

math.NA↗

arXiv2026

Viewpoint Recommendation for Point Cloud Labeling through Interaction Cost Modeling

Semantic segmentation of 3D point clouds is important for many applications, such as autonomous driving. To train semantic segmentation models, labeled point cloud segmentation datasets are essential. Meanwhile, point cloud labeling is time-consuming for annotators, which typically involves tuning the camera viewpoint and selecting points by lasso. To reduce the time cost of point cloud labeling, we propose a viewpoint recommendation approach to reduce annotators' labeling time costs. We adapt Fitts' law to model the time cost of lasso selection in point clouds. Using the modeled time cost, the viewpoint that minimizes the lasso selection time cost is recommended to the annotator. We build a data labeling system for semantic segmentation of 3D point clouds that integrates our viewpoint recommendation approach. The system enables users to navigate to recommended viewpoints for efficient annotation. Through an ablation study, we observed that our approach effectively reduced the data labeling time cost. We also qualitatively compare our approach with previous viewpoint selection approaches on different datasets.

cs.HC↗

arXiv2025

Sustainable Development Goals in Psychology: A Century of Progress in Publications

The Sustainable Development Goals (SDGs) offer a lens for tracking societal change, yet contributions from the social and behavioral sciences have rarely been integrated into policy agendas. To take stock and create a baseline and benchmark for the future, we assemble 233,061 psychology publications (1894 -- 2022) and tag them to the 17 SDGs using a query-based classifier. Health, education, work, inequality, and gender dominate the study of SDGs in psychology, shifting from an early focus on work to education and inequality, and since the 1960s, health. United States-based research leads across most goals. Other countries set distinct priorities (e.g., China: education and work; Australia: health). Women comprise about one-third of authors, concentrated in social and health goals, but have been underrepresented in STEM-oriented goals. The 2015 launch of the SDGs marked a turning point: SDG-tagged publications have been receiving more citations than comparable non-SDG work, reversing a pre-2015 deficit. Tracking the SDGs through psychology clarifies long-run engagement with social priorities, identifies evidence gaps, and guides priorities to accelerate the field's contribution to the SDG agenda.

cs.DL↗

arXiv2025

ALARM: Automated MLLM-Based Anomaly Detection in Complex-EnviRonment Monitoring with Uncertainty Quantification

The advance of Large Language Models (LLMs) has greatly stimulated research interest in developing multi-modal LLM (MLLM)-based visual anomaly detection (VAD) algorithms that can be deployed in complex environments. The challenge is that in these complex environments, the anomalies are sometimes highly contextual and also ambiguous, and thereby, uncertainty quantification (UQ) is a crucial capacity for an MLLM-based VAD system to succeed. In this paper, we introduce our UQ-supported MLLM-based VAD framework called ALARM. ALARM integrates UQ with quality-assurance techniques like reasoning chain, self-reflection, and MLLM ensemble for robust and accurate performance and is designed based on a rigorous probabilistic inference pipeline and computational process. Extensive empirical evaluations are conducted using the real-world smart-home benchmark data and wound image classification data, which shows ALARM's superior performance and its generic applicability across different domains for reliable decision-making.

cs.LG↗

arXiv2025

Constraining gravity with the decay rate of cosmological gravitational potential

A key task in cosmology is to test the validity of general relativity (GR) at cosmological scales and, therefore, to distinguish between dark energy and modified gravity (MG) as the driver of the late-time cosmic acceleration. The decay rate ($DR$) of cosmological gravitational potential, being sensitive to gravity and being immune to various astrophysical uncertainties, enables GR tests independent to other structure growth probes. Recently we have measured $DR$ at $0.2\leq z\leq 1.4$, combining the DR9 galaxy catalog from the DESI imaging surveys and Planck cosmic microwave background maps \citep{arXiv:2411.12594}. Here we use this measurement to test gravity, and restrict the analysis to one-parameter extensions to the standard $\Lambda$CDM cosmology. We consider four one-parameter MG parameterizations. One is $f(a)=\Omega_m^\gamma(a)$. The other three adopt the gravitational slip parameter $\eta=1$ and consider variations in the effective gravitational constant $G_{\rm eff}/G$ with the parameterization $\Sigma(a)=\Sigma_\Lambda \Omega_\Lambda(a)/\Omega_\Lambda$, $\Sigma(a)=\Sigma_1 a$ or $\Sigma(a)=\Sigma_2 a^2$. We find $\gamma=0.47^{+0.22}_{-0.15}$, consistent with the GR prediction $\gamma\simeq 0.55$. We also find $\Sigma_\Lambda=0.018^{+0.052}_{-0.053}$, $\Sigma_1=0.020^{+0.065}_{-0.062}$, and $\Sigma_2=0.027^{+0.067}_{-0.069}$, fully consistent with the GR case of $\Sigma=0$, regardless of parameterizations of $\Sigma(a)$. The constraining power is already competitive, while a factor of 2 further improvement is expected for the upcoming full-sky galaxy surveys.

astro-ph.CO↗

arXiv2025

Robustness Analysis and Controller Design of Arm-locking System in Space-based Gravitational Wave Detectors

Arm-locking frequency stabilization is a key technique for suppressing laser frequency noise in space-based gravitational-wave detectors. The robustness of the arm-locking control loop is crucial for maintaining laser frequency stability, which directly impacts the accuracy of gravitational-wave measurements. In this work, a parametric stability analysis framework is developed by combining the D-subdivision theory with the Semi-Discretization method to map the stability regions of arm-locking systems in the parameter space and identify their critical stability boundaries. Based on the frequency-domain characteristics, a robust arm-locking controller is designed to enhance loop stability under parameter perturbations. Theoretical analysis and time-domain simulations confirm that the proposed controller maintains closed-loop stability and realize suppression of laser frequency noise against parameter perturbation.

gr-qc↗

arXiv2025

Mapping the gender attrition gap in academic psychology

Women comprise the majority of students and early-career scholars in psychology, yet they are less likely to remain active in research over time. This pattern raises a central question: At what stages of academic careers do women disproportionately leave academia, and what factors drive their attrition? Using large-scale bibliometric data tracking 78,216 psychologists who began publishing between 2000 and 2014, we examine gender differences in research career attrition operationalized through publishing activity across the full trajectory from entry onward. Although women accounted for more than 60\% of new entrants, they experienced higher attrition rates than men, with the gender gap peaking approximately five years after first publication. Early-career performance, particularly first-authored publications, was the strongest predictor of subsequent retention, whereas last-authored publications were most closely associated with continued activity at later career stages. Collaboration patterns and institutional context also shaped career persistence, though less strongly than publication indicators. Notably, gender differences in research attrition persisted even after accounting for these career determinants, especially during early career stages. These findings suggest that gender inequality in psychology is driven less by recruitment than by differential retention over time. Addressing early-career vulnerability may therefore be essential to achieving equitable representation in senior academic leadership within the discipline.

cs.SI↗

arXiv2025

SYNBUILD-3D: A large, multi-modal, and semantically rich synthetic dataset of 3D building models at Level of Detail 4

3D building models are critical for applications in architecture, energy simulation, and navigation. Yet, generating accurate and semantically rich 3D buildings automatically remains a major challenge due to the lack of large-scale annotated datasets in the public domain. Inspired by the success of synthetic data in computer vision, we introduce SYNBUILD-3D, a large, diverse, and multi-modal dataset of over 6.2 million synthetic 3D residential buildings at Level of Detail (LoD) 4. In the dataset, each building is represented through three distinct modalities: a semantically enriched 3D wireframe graph at LoD 4 (Modality I), the corresponding floor plan images (Modality II), and a LiDAR-like roof point cloud (Modality III). The semantic annotations for each building wireframe are derived from the corresponding floor plan images and include information on rooms, doors, and windows. Through its tri-modal nature, future work can use SYNBUILD-3D to develop novel generative AI algorithms that automate the creation of 3D building models at LoD 4, subject to predefined floor plan layouts and roof geometries, while enforcing semantic-geometric consistency. Dataset and code samples are publicly available at https://github.com/kdmayer/SYNBUILD-3D.

cs.CV↗

arXiv2025

SmartHome-Bench: A Comprehensive Benchmark for Video Anomaly Detection in Smart Homes Using Multi-Modal Large Language Models

Video anomaly detection (VAD) is essential for enhancing safety and security by identifying unusual events across different environments. Existing VAD benchmarks, however, are primarily designed for general-purpose scenarios, neglecting the specific characteristics of smart home applications. To bridge this gap, we introduce SmartHome-Bench, the first comprehensive benchmark specially designed for evaluating VAD in smart home scenarios, focusing on the capabilities of multi-modal large language models (MLLMs). Our newly proposed benchmark consists of 1,203 videos recorded by smart home cameras, organized according to a novel anomaly taxonomy that includes seven categories, such as Wildlife, Senior Care, and Baby Monitoring. Each video is meticulously annotated with anomaly tags, detailed descriptions, and reasoning. We further investigate adaptation methods for MLLMs in VAD, assessing state-of-the-art closed-source and open-source models with various prompting techniques. Results reveal significant limitations in the current models' ability to detect video anomalies accurately. To address these limitations, we introduce the Taxonomy-Driven Reflective LLM Chain (TRLC), a new LLM chaining framework that achieves a notable 11.62% improvement in detection accuracy. The benchmark dataset and code are publicly available at https://github.com/Xinyi-0724/SmartHome-Bench-LLM.

cs.CV↗

arXiv2025

Manipulate intrinsic light-matter interaction with bound state in the continuum in van der Waals metasurfaces by artificial etching

The recent demonstrations of van der Waals (vdW) nanophotonics have opened new pathways for manipulating the light-matter interaction in an intrinsic manner, leading to fascinating achievements in tunable magneto-optics by self-hybrid polaritons, indirect bandgap lasering, and exceptionally enhanced optical nonlinearity. However, the anisotropic atomic lattice, chemically active side walls, and distinct enthalpies of formation across vdW materials, pose significant challenges in nanofabrication and material choices, hindering the realization of high-Q resonant mode on arbitrary materials. In this work, we propose an etch-free vdW structure that mimics the shallow etching, termed "artificial etching". This approach utilizes a low refractive index (LRI) perturbation layer made of photoresist, drastically reducing radiation loss and experimentally achieving a remarkable Q factor of up to 348, which is comparable to the highest values reported in vdW nanophotonics. We demonstrate room-temperature polaritons in etch-free structures using four representative materials (WS$_2$, MoS$_2$, WSe$_2$, and MoSe$_2$) through self-hybridization of high-Q (quasi-)bound states in the continuum (BIC) modes and excitons, achieving a Rabi-splitting of approximately 80 meV, which significantly surpasses the intrinsic excitonic loss. Furthermore, we showcase optical modulation of indirect bandgap emission in bulk WS$_2$ and direct exciton emission in heterostructures, achieving substantial polarization-dependent enhancement of their emission efficiencies. The proposed etch-free vdW structure provides a versatile platform for high-Q nanophononics while preserving material integrity, advancing applications in photoelectronic and quantum devices.

physics.optics↗

arXiv2024

A Distributionally Robust Optimization Framework for Stochastic Assessment of Power System Flexibility in Economic Dispatch

Given the complexity of power systems, particularly the high-dimensional variability of net loads, accurately depicting the entire operational range of net loads poses a challenge. To address this, recent methodologies have sought to gauge the maximum range of net load uncertainty across all buses. In this paper, we consider the stochastic nature of the net load and introduce a distributionally robust optimization framework that assesses system flexibility stochastically, accommodating a minimal extent of system violations. We verify the proposed method by solving the flexibility of the real-time economic dispatch problem on four IEEE standard test systems. Compared to traditional deterministic flexibility evaluations, our approach consistently yields less conservative flexibility outcomes.

math.OC↗

arXiv2024

Just read twice: closing the recall gap for recurrent language models

Recurrent large language models that compete with Transformers in language modeling perplexity are emerging at a rapid rate (e.g., Mamba, RWKV). Excitingly, these architectures use a constant amount of memory during inference. However, due to the limited memory, recurrent LMs cannot recall and use all the information in long contexts leading to brittle in-context learning (ICL) quality. A key challenge for efficient LMs is selecting what information to store versus discard. In this work, we observe the order in which information is shown to the LM impacts the selection difficulty. To formalize this, we show that the hardness of information recall reduces to the hardness of a problem called set disjointness (SD), a quintessential problem in communication complexity that requires a streaming algorithm (e.g., recurrent model) to decide whether inputted sets are disjoint. We empirically and theoretically show that the recurrent memory required to solve SD changes with set order, i.e., whether the smaller set appears first in-context. Our analysis suggests, to mitigate the reliance on data order, we can put information in the right order in-context or process prompts non-causally. Towards that end, we propose: (1) JRT-Prompt, where context gets repeated multiple times in the prompt, effectively showing the model all data orders. This gives $11.0 \pm 1.3$ points of improvement, averaged across $16$ recurrent LMs and the $6$ ICL tasks, with $11.9\times$ higher throughput than FlashAttention-2 for generation prefill (length $32$k, batch size $16$, NVidia H100). We then propose (2) JRT-RNN, which uses non-causal prefix-linear-attention to process prompts and provides $99\%$ of Transformer quality at $360$M params., $30$B tokens and $96\%$ at $1.3$B params., $50$B tokens on average across the tasks, with $19.2\times$ higher throughput for prefill than FA2.

cs.CL↗

arXiv2024

LEVA: Using Large Language Models to Enhance Visual Analytics

Visual analytics supports data analysis tasks within complex domain problems. However, due to the richness of data types, visual designs, and interaction designs, users need to recall and process a significant amount of information when they visually analyze data. These challenges emphasize the need for more intelligent visual analytics methods. Large language models have demonstrated the ability to interpret various forms of textual data, offering the potential to facilitate intelligent support for visual analytics. We propose LEVA, a framework that uses large language models to enhance users' VA workflows at multiple stages: onboarding, exploration, and summarization. To support onboarding, we use large language models to interpret visualization designs and view relationships based on system specifications. For exploration, we use large language models to recommend insights based on the analysis of system status and data to facilitate mixed-initiative exploration. For summarization, we present a selective reporting strategy to retrace analysis history through a stream visualization and generate insight reports with the help of large language models. We demonstrate how LEVA can be integrated into existing visual analytics systems. Two usage scenarios and a user study suggest that LEVA effectively aids users in conducting visual analytics.

cs.HC↗

arXiv2023

Grid-Aware On-Route Fast-Charging Infrastructure Planning for Battery Electric Bus with Equity Considerations: A Case Study in South King County

The transition from traditional bus fleets to zero-emission ones necessitates the development of effective planning models for battery electric bus (BEB) charging infrastructure. On-route fast charging stations, distinct from on-base charging stations, present unique challenges related to safe operation and power supply capacity, making it difficult to control grid operational costs. This paper establishes a novel framework that integrates the bus route network and power network, which leverages the inter-dependency between both networks to optimize the planning outcomes of on-route BEB charging stations in South King County. The problem is formulated as a mixed-integer second-order cone programming model, aiming to minimize the overall planning cost, which includes investments in charging equipment, power facility, and grid operation. Furthermore, fairness measurements are incorporated into the planning process, allowing for the consideration of both horizontal transit equity and vertical transit equity based on different zone merging criteria within the county's existing census tracts. The results of this planning model offer valuable insights into achieving both economic efficiency and social justice in the design of on-route charging facilities for BEBs in South King County.

eess.SY↗