arXiv subjects

Ruoxi Wang

Publications and source records attributed to Ruoxi Wang.

At least 19 recordsLinked to original sources

arXiv2026

The Efficiency Gap in Byte Modeling

Modern language models have historically relied on two dominant design choices: subword tokenization and autoregressive (AR) ordering. These design decisions bake in priors that dictate a model's learning. Recently, two alternative paradigms have challenged this: byte-level modeling, which bypasses static statistically-derived token vocabularies, and masked diffusion modeling (MDM), which conducts parallel, non-sequential generation. Their intersection represents a fully end-to-end modality-agnostic generative prototype; however, removing these structural priors incurs a significant computational cost. In this work, we investigate this cost through a compute-matched scaling study. Our results reveal that the performance penalty of byte modeling is not uniform; across scale, the scaling overhead of byte modeling is worse for MDM than for AR. We hypothesize that this disparity stems from context fragility: while AR's stable causal history allows models to naturally rediscover subword patterns, the MDM objective destroys the local contiguity required to efficiently resolve semantics from raw bytes. Our findings from controlled permutation experiments suggest that future modality-agnostic designs must incorporate alternative structural biases to maintain viable scaling trajectories in the byte regime.

cs.LG↗

arXiv2025

ProSoftArena: Benchmarking Hierarchical Capabilities of Multimodal Agents in Professional Software Environments

Multimodal agents are making rapid progress on general computer-use tasks, yet existing benchmarks remain largely confined to browsers and basic desktop applications, falling short in professional software workflows that dominate real-world scientific and industrial practice. To close this gap, we introduce ProSoftArena, a benchmark and platform specifically for evaluating multimodal agents in professional software environments. We establish the first capability hierarchy tailored to agent use of professional software and construct a benchmark of 436 realistic work and research tasks spanning 6 disciplines and 13 core professional applications. To ensure reliable and reproducible assessment, we build an executable real-computer environment with an execution-based evaluation framework and uniquely incorporate a human-in-the-loop evaluation paradigm. Extensive experiments show that even the best-performing agent attains only a 24.4\% success rate on L2 tasks and completely fails on L3 multi-software workflow. In-depth analysis further provides valuable insights for addressing current agent limitations and more effective design principles, paving the way to build more capable agents in professional software settings. This project is available at: https://prosoftarena.github.io.

cs.SE↗

arXiv2025

Micro-Electro-Mechanical System Vapor Cells With Passivated Internal Cavities

Micro-Electro-Mechanical, so called 'MEMs,' vapor cells are a key component in atom-based quantum sensors, such as clocks, gyroscopes, electric field sensors and magnetometers. MEMs vapor cell fabrication for Rydberg atom radio frequency sensors is particularly demanding. The Rydberg states used for the sensor can shift in a constant electric field which can be generated by the internal surfaces of the vapor cell cavity. The ratio of the detection wavelength to vapor cell size can span a large range, meaning that the radio frequency field-vapor cell interaction is a critical design consideration. In many radio frequency sensing cases, there is a desire to minimize the interaction between the vapor cell and the target radio frequency field, as well as assure that every vapor cell behaves uniformly. These criterion favor MEMs vapor cells with low background electric fields. Known inert, organic coatings cannot survive the bonding temperatures required for conventional anodic bonding of a MEMs vapor cell. Applying inert, organic coatings to the internal cavities of MEMs vapor cells is a longstanding challenge. In this paper, we present a low temperature bonding scheme that is compatible with coating the internal cavity of a MEMs vapor cell with Octadecyltrichlorosilane (CH$_3\,$(CH$_2$)$_{17}\,$SiCl$_3$, OTS). The coating prevents the Cs used in the vapor cell from sticking to the walls. Spectral linewidths of $\sim300\,$kHz are obtained using Rydberg spectroscopy, with energy shifts corresponding to electric fields $<$10$\,$mV$\,$cm$^{-1}$.

physics.atom-ph↗

arXiv2025

LIFT: Automating Symbolic Execution Optimization with Large Language Models for AI Networks

Dynamic Symbolic Execution (DSE) is a key technique in program analysis, widely used in software testing, vulnerability discovery, and formal verification. In distributed AI systems, DSE plays a crucial role in identifying hard-to-detect bugs, especially those arising from complex network communication patterns. However, traditional approaches to symbolic execution are often hindered by scalability issues and inefficiencies, particularly in large-scale systems. This paper introduces LIFT (Large-language-model Integrated Functional-equivalent-IR Transformation), a novel framework that leverages Large Language Models (LLMs) to automate the optimization of Intermediate Representations (IRs) in symbolic execution. LIFT addresses the challenges of symbolic execution by providing a scalable, context-sensitive solution for IR transformation. The framework consists of two phases: IR Analysis and Optimization, where LLMs optimize time-intensive IR blocks, and Symbolic Execution and Validation, which includes benchmarking and semantic verification to ensure correctness and generalizability. Experiments on real-world binaries demonstrated significant performance improvements, including a 53.5\% reduction in execution time for bigtest and a 10.24\% reduction for random, along with reductions in IR statements, PUT instructions, and temporary variables. These results demonstrate that LLMs simplify IRs while maintaining functional correctness, enhancing symbolic execution in distributed AI systems.

cs.CR↗

arXiv2025

All-optical radio-frequency phase detection for Rydberg atom sensors using oscillatory dynamics

Rydberg atom radio frequency sensors are a unique platform for precision electromagnetic field measurement, e.g. they have extraordinary carrier bandwidth spanning MHz-THz and can be self-calibrated. These photonic sensors use lasers to prepare and read out the atomic response to a radio frequency electromagnetic field. Most work on Rydberg atom sensors centers on radio frequency electric field strength because the sensor functions as a square law detector, unless an external radio frequency heterodyning field is used. A heterodyning field acts as a local oscillator and enables phase read out at the expense of the radio frequency equipment necessary to generate it. In order to overcome the disadvantages of a radio frequency local oscillator, we investigate all-optical phase-sensitive detection using a five-level closed-loop excitation scheme. We show that under finite detuning of the loop fields, the atomic response oscillates at the frequency of the detuning. The oscillation is transferred to a probe laser absorption signal. The phase, frequency and amplitude of the radio frequency signal are imprinted on the oscillatory dynamics and can be determined using demodulation and matched filter techniques applied to the probe laser transmission signal.

physics.atom-ph↗

arXiv2025

PsychBench: A comprehensive and professional benchmark for evaluating the performance of LLM-assisted psychiatric clinical practice

The advent of Large Language Models (LLMs) offers potential solutions to address problems such as shortage of medical resources and low diagnostic consistency in psychiatric clinical practice. Despite this potential, a robust and comprehensive benchmarking framework to assess the efficacy of LLMs in authentic psychiatric clinical environments is absent. This has impeded the advancement of specialized LLMs tailored to psychiatric applications. In response to this gap, by incorporating clinical demands in psychiatry and clinical data, we proposed a benchmarking system, PsychBench, to evaluate the practical performance of LLMs in psychiatric clinical settings. We conducted a comprehensive quantitative evaluation of 16 LLMs using PsychBench, and investigated the impact of prompt design, chain-of-thought reasoning, input text length, and domain-specific knowledge fine-tuning on model performance. Through detailed error analysis, we identified strengths and potential limitations of the existing models and suggested directions for improvement. Subsequently, a clinical reader study involving 60 psychiatrists of varying seniority was conducted to further explore the practical benefits of existing LLMs as supportive tools for psychiatrists of varying seniority. Through the quantitative and reader evaluation, we show that while existing models demonstrate significant potential, they are not yet adequate as decision-making tools in psychiatric clinical practice. The reader study further indicates that, as an auxiliary tool, LLM could provide particularly notable support for junior psychiatrists, effectively enhancing their work efficiency and overall clinical quality. To promote research in this area, we will make the dataset and evaluation framework publicly available, with the hope of advancing the application of LLMs in psychiatric clinical settings.

cs.CL↗

arXiv2025

Hierarchical Contextual Manifold Alignment for Structuring Latent Representations in Large Language Models

The organization of latent token representations plays a crucial role in determining the stability, generalization, and contextual consistency of language models, yet conventional approaches to embedding refinement often rely on parameter modifications that introduce additional computational overhead. A hierarchical alignment method was introduced to restructure token embeddings without altering core model weights, ensuring that representational distributions maintained coherence across different linguistic contexts. Experimental evaluations demonstrated improvements in rare token retrieval, adversarial robustness, and long-range dependency tracking, highlighting the advantages of hierarchical structuring in mitigating inconsistencies in latent space organization. The comparative analysis against conventional fine-tuning and embedding perturbation methods revealed that hierarchical restructuring maintained computational efficiency while achieving measurable gains in representation quality. Structural refinements introduced through the alignment process resulted in improved contextual stability across varied linguistic tasks, reducing inconsistencies in token proximity relationships and enhancing interpretability in language generation. A detailed computational assessment confirmed that the realignment process introduced minimal inference overhead, ensuring that representational improvements did not compromise model efficiency. The findings reinforced the broader significance of structured representation learning, illustrating that hierarchical embedding modifications could serve as an effective strategy for refining latent space distributions while preserving pre-learned semantic associations.

cs.CL↗

arXiv2024

AutoIoT: Automated IoT Platform Using Large Language Models

IoT platforms, particularly smart home platforms providing significant convenience to people's lives such as Apple HomeKit and Samsung SmartThings, allow users to create automation rules through trigger-action programming. However, some users may lack the necessary knowledge to formulate automation rules, thus preventing them from fully benefiting from the conveniences offered by smart home technology. To address this, smart home platforms provide pre-defined automation policies based on the smart home devices registered by the user. Nevertheless, these policies, being pre-generated and relatively simple, fail to adequately cover the diverse needs of users. Furthermore, conflicts may arise between automation rules, and integrating conflict detection into the IoT platform increases the burden on developers. In this paper, we propose AutoIoT, an automated IoT platform based on Large Language Models (LLMs) and formal verification techniques, designed to achieve end-to-end automation through device information extraction, LLM-based rule generation, conflict detection, and avoidance. AutoIoT can help users generate conflict-free automation rules and assist developers in generating codes for conflict detection, thereby enhancing their experience. A code adapter has been designed to separate logical reasoning from the syntactic details of code generation, enabling LLMs to generate code for programming languages beyond their training data. Finally, we evaluated the performance of AutoIoT and presented a case study demonstrating how AutoIoT can integrate with existing IoT platforms.

cs.CR↗

arXiv2024

I'm Spartacus, No, I'm Spartacus: Measuring and Understanding LLM Identity Confusion

Large Language Models (LLMs) excel in diverse tasks such as text generation, data analysis, and software development, making them indispensable across domains like education, business, and creative industries. However, the rapid proliferation of LLMs (with over 560 companies developing or deploying them as of 2024) has raised concerns about their originality and trustworthiness. A notable issue, termed identity confusion, has emerged, where LLMs misrepresent their origins or identities. This study systematically examines identity confusion through three research questions: (1) How prevalent is identity confusion among LLMs? (2) Does it arise from model reuse, plagiarism, or hallucination? (3) What are the security and trust-related impacts of identity confusion? To address these, we developed an automated tool combining documentation analysis, self-identity recognition testing, and output similarity comparisons--established methods for LLM fingerprinting--and conducted a structured survey via Credamo to assess its impact on user trust. Our analysis of 27 LLMs revealed that 25.93% exhibit identity confusion. Output similarity analysis confirmed that these issues stem from hallucinations rather than replication or reuse. Survey results further highlighted that identity confusion significantly erodes trust, particularly in critical tasks like education and professional use, with declines exceeding those caused by logical errors or inconsistencies. Users attributed these failures to design flaws, incorrect training data, and perceived plagiarism, underscoring the systemic risks posed by identity confusion to LLM reliability and trustworthiness.

cs.CR↗

arXiv2024

Dynamical control in a prethermalized molecular ultracold plasma: Local dissipation drives global relaxation

Prethermalization occurs as an important phase in the dynamics of many-body systems when strong coupling drives a quasi-equilibrium in a subspace separated from the thermodynamic equilibrium by the restriction of a gap in energy or other conserved quantity. Here, we report the signature of an enduring prethermal regime of arrested relaxation in the molecular ultracold plasma that forms following the avalanche of a state-selected Rydberg gas of nitric oxide. Electron collisions mix orbital angular momentum, scattering Rydberg molecules to states of very high-$\ell$. Spontaneous predissociation purifies this non-penetrating character, creating an extraordinary gap between the plasma states of $n \approx \ell$, with measured $n>200$ and penetrating states of $\ell = 0, ~1$ and 2. Evolution to a statistically equilibrated state of N and O atoms cannot occur without Rydberg electron penetration, and this gap blocks relaxation for a millisecond or more. Evolving through the critical phase, electrons that balance the NO$^+$ charge behave as though localized in the prethermal phase and play an ineffective role in bridging this gap. However, the application of a weak radiofrequency (RF) field promotes a dramatic degree of relaxation owing to electron collisions. On an entirely different scale, exciting a quantum-state transition in an exceedingly small fraction of the molecules in the prethermalized ensemble acts with even greater effect to drive the entire system toward equilibrium. We ascribe this to dissipative character added to a small fraction of the states in the prethermally localized ensemble. Using the Lindblad master equation, we illustrate qualitatively similar dynamics for a toy model of an open quantum system that consists of a localized set of spins on which dissipation acts locally at a single site.

cond-mat.quant-gas↗

arXiv2023

Hiformer: Heterogeneous Feature Interactions Learning with Transformers for Recommender Systems

Learning feature interaction is the critical backbone to building recommender systems. In web-scale applications, learning feature interaction is extremely challenging due to the sparse and large input feature space; meanwhile, manually crafting effective feature interactions is infeasible because of the exponential solution space. We propose to leverage a Transformer-based architecture with attention layers to automatically capture feature interactions. Transformer architectures have witnessed great success in many domains, such as natural language processing and computer vision. However, there has not been much adoption of Transformer architecture for feature interaction modeling in industry. We aim at closing the gap. We identify two key challenges for applying the vanilla Transformer architecture to web-scale recommender systems: (1) Transformer architecture fails to capture the heterogeneous feature interactions in the self-attention layer; (2) The serving latency of Transformer architecture might be too high to be deployed in web-scale recommender systems. We first propose a heterogeneous self-attention layer, which is a simple yet effective modification to the self-attention layer in Transformer, to take into account the heterogeneity of feature interactions. We then introduce \textsc{Hiformer} (\textbf{H}eterogeneous \textbf{I}nteraction Trans\textbf{former}) to further improve the model expressiveness. With low-rank approximation and model pruning, \hiformer enjoys fast inference for online deployment. Extensive offline experiment results corroborates the effectiveness and efficiency of the \textsc{Hiformer} model. We have successfully deployed the \textsc{Hiformer} model to a real world large scale App ranking model at Google Play, with significant improvement in key engagement metrics (up to +2.66\%).

cs.IR↗

arXiv2023

Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN)

Industry recommender systems usually suffer from highly-skewed long-tail item distributions where a small fraction of the items receives most of the user feedback. This skew hurts recommender quality especially for the item slices without much user feedback. While there have been many research advances made in academia, deploying these methods in production is very difficult and very few improvements have been made in industry. One challenge is that these methods often hurt overall performance; additionally, they could be complex and expensive to train and serve. In this work, we aim to improve tail item recommendations while maintaining the overall performance with less training and serving cost. We first find that the predictions of user preferences are biased under long-tail distributions. The bias comes from the differences between training and serving data in two perspectives: 1) the item distributions, and 2) user's preference given an item. Most existing methods mainly attempt to reduce the bias from the item distribution perspective, ignoring the discrepancy from user preference given an item. This leads to a severe forgetting issue and results in sub-optimal performance. To address the problem, we design a novel Cross Decoupling Network (CDN) (i) decouples the learning process of memorization and generalization on the item side through a mixture-of-expert architecture; (ii) decouples the user samples from different distributions through a regularized bilateral branch network. Finally, a new adapter is introduced to aggregate the decoupled vectors, and softly shift the training attention to tail items. Extensive experimental results show that CDN significantly outperforms state-of-the-art approaches on benchmark datasets. We also demonstrate its effectiveness by a case study of CDN in a large-scale recommendation system at Google.

cs.IR↗

arXiv2023

HyperFormer: Learning Expressive Sparse Feature Representations via Hypergraph Transformer

Learning expressive representations for high-dimensional yet sparse features has been a longstanding problem in information retrieval. Though recent deep learning methods can partially solve the problem, they often fail to handle the numerous sparse features, particularly those tail feature values with infrequent occurrences in the training data. Worse still, existing methods cannot explicitly leverage the correlations among different instances to help further improve the representation learning on sparse features since such relational prior knowledge is not provided. To address these challenges, in this paper, we tackle the problem of representation learning on feature-sparse data from a graph learning perspective. Specifically, we propose to model the sparse features of different instances using hypergraphs where each node represents a data instance and each hyperedge denotes a distinct feature value. By passing messages on the constructed hypergraphs based on our Hypergraph Transformer (HyperFormer), the learned feature representations capture not only the correlations among different instances but also the correlations among features. Our experiments demonstrate that the proposed approach can effectively improve feature representation learning on sparse features.

cs.IR↗

arXiv2023

Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems

Learning high-quality feature embeddings efficiently and effectively is critical for the performance of web-scale machine learning systems. A typical model ingests hundreds of features with vocabularies on the order of millions to billions of tokens. The standard approach is to represent each feature value as a d-dimensional embedding, introducing hundreds of billions of parameters for extremely high-cardinality features. This bottleneck has led to substantial progress in alternative embedding algorithms. Many of these methods, however, make the assumption that each feature uses an independent embedding table. This work introduces a simple yet highly effective framework, Feature Multiplexing, where one single representation space is used across many different categorical features. Our theoretical and empirical analysis reveals that multiplexed embeddings can be decomposed into components from each constituent feature, allowing models to distinguish between features. We show that multiplexed representations lead to Pareto-optimal parameter-accuracy tradeoffs for three public benchmark datasets. Further, we propose a highly practical approach called Unified Embedding with three major benefits: simplified feature configuration, strong adaptation to dynamic data distributions, and compatibility with modern hardware. Unified embedding gives significant improvements in offline and online metrics compared to highly competitive baselines across five web-scale search, ads, and recommender systems, where it serves billions of users across the world in industry-leading products.

cs.LG↗

arXiv2023

High-dimensional Automated Radiation Therapy Treatment Planning via Bayesian Optimization

Radiation therapy treatment planning can be viewed as an iterative hyperparameter tuning process to balance conflicting clinical goals. In this work, we investigated the performance of modern Bayesian Optimization (BO) methods on automated treatment planning problems in high-dimensional settings. 20 locally advanced rectal cancer patients treated with intensity-modulated radiation therapy (IMRT) were retrospectively selected as test cases. We implemented an automated treatment planning framework that adjusts dose objectives and weights simultaneously, and tested the performance of two BO methods on the treatment planning task: one standard BO method (GPEI) and one BO method dedicated to high-dimensional problems (SAAS-BO). A random tuning method was also included as the baseline. We compared and analyzed the three automated methods' plan quality and planning efficiency and the different search patterns of the two BO methods. For the target structures, the SAAS-BO plans achieved comparable hot spot control ($p=0.43$) and homogeneity ($p=0.96$) with the clinical plans, significantly better than the GPEI and random plans ($p<0.05$). Both SAAS-BO and GPEI plans significantly outperformed the clinical plans in conformity and dose spillage ($p<0.05$). Compared with clinical plans, the treatment plans generated by the three automated methods all made reductions in evaluated dosimetric indices for the femoral head and the bladder. The analysis of the underlying predictive models has shown that both BO procedures have identified similar important planning parameters. This work implemented a BO-based hyperparameter tuning framework for automated treatment planning. Both tested BO methods were able to produce high-quality treatment plans and the model analysis also confirmed the intrinsic low dimensionality of the tested treatment planning problems.

physics.med-ph↗

arXiv2022

Artificial Intelligence for Scientific Research: Authentic Research Education Framework

We report a framework that enables the wide adoption of authentic research educational methodology at various schools by addressing common barriers. The guiding principles we present were applied to implement a program in which teams of students with complementary skills develop useful artificial intelligence (AI) solutions for researchers in natural sciences. To accomplish this, we work with research laboratories that reveal/specify their needs, and then our student teams work on the discovery, design, and development of an AI solution for unique problems using a consulting-like arrangement. To date, our group has been operating at New York University (NYU) for seven consecutive semesters, has engaged more than a hundred students, ranging from first-year college students to master's candidates, and has worked with more than twenty projects and collaborators. While creating education benefits for students, our approach also directly benefits scientists, who get an opportunity to evaluate the usefulness of machine learning for their specific needs.

cs.CY↗

arXiv2021

Block Basis Factorization for Scalable Kernel Matrix Evaluation

Kernel methods are widespread in machine learning; however, they are limited by the quadratic complexity of the construction, application, and storage of kernel matrices. Low-rank matrix approximation algorithms are widely used to address this problem and reduce the arithmetic and storage cost. However, we observed that for some datasets with wide intra-class variability, the optimal kernel parameter for smaller classes yields a matrix that is less well approximated by low-rank methods. In this paper, we propose an efficient structured low-rank approximation method -- the Block Basis Factorization (BBF) -- and its fast construction algorithm to approximate radial basis function (RBF) kernel matrices. Our approach has linear memory cost and floating-point operations for many machine learning kernels. BBF works for a wide range of kernel bandwidth parameters and extends the domain of applicability of low-rank approximation methods significantly. Our empirical results demonstrate the stability and superiority over the state-of-art kernel approximation algorithms.

stat.ML↗

arXiv2021

Surgical Gesture Recognition Based on Bidirectional Multi-Layer Independently RNN with Explainable Spatial Feature Extraction

Minimally invasive surgery mainly consists of a series of sub-tasks, which can be decomposed into basic gestures or contexts. As a prerequisite of autonomic operation, surgical gesture recognition can assist motion planning and decision-making, and build up context-aware knowledge to improve the surgical robot control quality. In this work, we aim to develop an effective surgical gesture recognition approach with an explainable feature extraction process. A Bidirectional Multi-Layer independently RNN (BML-indRNN) model is proposed in this paper, while spatial feature extraction is implemented via fine-tuning of a Deep Convolutional Neural Network(DCNN) model constructed based on the VGG architecture. To eliminate the black-box effects of DCNN, Gradient-weighted Class Activation Mapping (Grad-CAM) is employed. It can provide explainable results by showing the regions of the surgical images that have a strong relationship with the surgical gesture classification results. The proposed method was evaluated based on the suturing task with data obtained from the public available JIGSAWS database. Comparative studies were conducted to verify the proposed framework. Results indicated that the testing accuracy for the suturing task based on our proposed method is 87.13%, which outperforms most of the state-of-the-art algorithms.

cs.CV↗