arXiv subjects

Tam Nguyen

Publications and source records attributed to Tam Nguyen.

At least 19 recordsLinked to original sources

arXiv2026

AI-accelerated End-to-End Framework for Rapid Professional Upskilling

By 2030, 59 of every 100 workers will need reskilling or upskilling, yet the average time to close an enterprise skills gap grew from roughly 3 days in 2014 to 36 days in 2018. Most current frameworks accelerate single stages of upskilling programs and generally lack industry validation. We present an end-to-end framework that applies AI acceleration across five stages of knowledge acquisition, content development, content review and verification, teaching, and assessment development; with a strong focus on both production and learning efficiency. Three strong external signals validates the framework: the US National Association of State Boards of Accountancy reviewed and approved an upskilling program built on the framework for continuing-professional-education credits; 3 learners followed the program and passed the NVIDIA Certified Professional in Agentic AI exam in a significantly short amount of time, with 14 more in progress; the program's knowledge base supports complex downstream analysis such as the production of a robust 1,267 risk item dataset for managing multi-agent AI system risks.

cs.AI↗

arXiv2026

Hybrid Improper Ferroelectricity and Moir\'e Superlattices-induced Exciton Quantization in Layered 2D Halide Perovskite

2D Ruddlesden-Popper perovskites are compelling platforms for quantum-confined optoelectronics. However, polar order in iodide composition remains rare under ambient conditions, and the mechanistic origin of anomalous photoluminescence in this class of perovskite is still speculative. Here, we demonstrate that solution-grown $(PA)_2FAPb_2I_7$ single crystals develop an inadvertent moir\'e superlattice through pseudo-merohedral twinning, driven by hybrid improper ferroelectricity in which trilinear mode coupling between two primary zone-boundary modes ($X_2^+$ and $X_3^-$) and a secondary $\Gamma_4^-$ polar displacement simultaneously breaks inversion symmetry and imposes a ca. 5.17{\deg} rotational misalignment between adjacent layers. This symmetry breaking activates one of the highest piezoelectric coefficients $d_{33}$ (ca. 20 pm/V) reported among 2D perovskites. This misalignment generates a moir\'e superlattice that undergoes a thermally driven commensurate-incommensurate transition, switching between a periodic confinement potential that quantizes excitons into an equidistant photoluminescence ladder at 123 K and a disordered incommensurate phase with broadened emission at 298 K. These emissions are attributed to moir\'e-confined excitons, resolving a longstanding debate on anomalous secondary photoluminescence in layered 2D perovskites and opening pathways to twistronics, photoferroelectrics and piezo-optoelectronic devices.

cond-mat.mtrl-sci↗

arXiv2026

Minimizing Collateral Damage in Activation Steering

Activation steering is a method for controlling Large Language Model (LLM) behavior by intervening in its internal representations to increase the alignment with a specific target feature direction. However, standard interventions, such as vector addition, often cause ``collateral damage", defined as unintended changes in the alignment of activations along other non-target feature directions. This damage occurs because standard methods implicitly assume the isotropy of non-target features. In this work, we provide a mathematical formalization of collateral damage and introduce a principled framework that models steering as a constrained optimization problem. Our method finds a new activation that minimizes the expected squared collateral change weighted by the empirical second-moment matrix of activations. This weighting encodes the nonuniform cost of the perturbation in different feature directions, in contrast to isotropic approaches that penalize changes uniformly in all feature directions. By accounting for the empirical second-moment of activations, our approach achieves more precise control while reducing the degradation of model performance on unrelated tasks.

cs.LG↗

arXiv2026

Security Considerations for Multi-agent Systems

Multi-agent artificial intelligence systems or MAS are systems of autonomous agents that exercise delegated tool authority, share persistent memory, and coordinate via inter-agent communication. MAS introduces qualitatively distinct security vulnerabilities from those documented for singular AI models. Existing security and governance frameworks were not designed for these emerging attack surfaces. This study systematically characterizes the threat landscape of MAS and quantitatively evaluates 16 security frameworks for AI against it. A four-phase methodology is proposed: constructing a deep technical knowledge base of production multi-agent architectures; conducting generative AI-assisted threat modeling scoped to MAS cybersecurity risks and validated by domain experts; structuring survey plans at individual-threat granularity; and scoring each framework on a three-point scale against the cybersecurity risks. The risks were organized into 193 distinct main threat items across nine risk categories. The expected minimal average score is 2. No reviewed framework achieves majority coverage of any single category. Non-Determinism (mean score 1.231 across all 16 frameworks) and Data Leakage (1.340) are the most under-addressed domains. The OWASP Agentic Security Initiative leads overall at 65.3\% coverage and in the design phase; the CDAO Generative AI Responsible AI Toolkit leads in development and operational coverage. These results provide the first empirical cross-framework comparison for MAS security and offer evidence-based guidance for framework selection. Please check back for information on the published journal version.

cs.CR↗

arXiv2025

OpenRoboCare: A Multimodal Multi-Task Expert Demonstration Dataset for Robot Caregiving

We present OpenRoboCare, a multimodal dataset for robot caregiving, capturing expert occupational therapist demonstrations of Activities of Daily Living (ADLs). Caregiving tasks involve complex physical human-robot interactions, requiring precise perception under occlusions, safe physical contact, and long-horizon planning. While recent advances in robot learning from demonstrations have shown promise, there is a lack of a large-scale, diverse, and expert-driven dataset that captures real-world caregiving routines. To address this gap, we collect data from 21 occupational therapists performing 15 ADL tasks on two manikins. The dataset spans five modalities: RGB-D video, pose tracking, eye-gaze tracking, task and action annotations, and tactile sensing, providing rich multimodal insights into caregiver movement, attention, force application, and task execution strategies. We further analyze expert caregiving principles and strategies, offering insights to improve robot efficiency and task feasibility. Additionally, our evaluations demonstrate that OpenRoboCare presents challenges for state-of-the-art robot perception and human activity recognition methods, both critical for developing safe and adaptive assistive robots, highlighting the value of our contribution. See our website for additional visualizations: https://emprise.cs.cornell.edu/robo-care/.

cs.RO↗

arXiv2025

Improving Routing in Sparse Mixture of Experts with Graph of Tokens

Sparse Mixture of Experts (SMoE) has emerged as a key to achieving unprecedented scalability in deep learning. By activating only a small subset of parameters per sample, SMoE achieves an exponential increase in parameter counts while maintaining a constant computational overhead. However, SMoE models are susceptible to routing fluctuations--changes in the routing of a given input to its target expert--at the late stage of model training, leading to model non-robustness. In this work, we unveil the limitation of SMoE through the perspective of the probabilistic graphical model (PGM). Through this PGM framework, we highlight the independence in the expert-selection of tokens, which exposes the model to routing fluctuation and non-robustness. Alleviating this independence, we propose the novel Similarity-Aware (S)MoE, which considers interactions between tokens during expert selection. We then derive a new PGM underlying an (S)MoE-Attention block, going beyond just a single (S)MoE layer. Leveraging the token similarities captured by the attention matrix, we propose the innovative Attention-Aware (S)MoE, which employs the attention matrix to guide the routing of tokens to appropriate experts in (S)MoE. We theoretically prove that Similarity/Attention-Aware routing help reduce the entropy of expert selection, resulting in more stable token routing mechanisms. We empirically validate our models on various tasks and domains, showing significant improvements in reducing routing fluctuations, enhancing accuracy, and increasing model robustness over the baseline MoE-Transformer with token routing via softmax gating.

cs.LG↗

arXiv2024

A Primal-Dual Framework for Transformers and Neural Networks

Self-attention is key to the remarkable success of transformers in sequence modeling tasks including many applications in natural language processing and computer vision. Like neural network layers, these attention mechanisms are often developed by heuristics and experience. To provide a principled framework for constructing attention layers in transformers, we show that the self-attention corresponds to the support vector expansion derived from a support vector regression problem, whose primal formulation has the form of a neural network layer. Using our framework, we derive popular attention layers used in practice and propose two new attentions: 1) the Batch Normalized Attention (Attention-BN) derived from the batch normalization layer and 2) the Attention with Scaled Head (Attention-SH) derived from using less training data to fit the SVR model. We empirically demonstrate the advantages of the Attention-BN and Attention-SH in reducing head redundancy, increasing the model's accuracy, and improving the model's efficiency in a variety of practical applications including image and time-series classification.

cs.LG↗

arXiv2024

PIDformer: Transformer Meets Control Theory

In this work, we address two main shortcomings of transformer architectures: input corruption and rank collapse in their output representation. We unveil self-attention as an autonomous state-space model that inherently promotes smoothness in its solutions, leading to lower-rank outputs and diminished representation capacity. Moreover, the steady-state solution of the model is sensitive to input perturbations. We incorporate a Proportional-Integral-Derivative (PID) closed-loop feedback control system with a reference point into the model to improve robustness and representation capacity. This integration aims to preserve high-frequency details while bolstering model stability, rendering it more noise-resilient. The resulting controlled state-space model is theoretically proven robust and adept at addressing the rank collapse. Motivated by this control framework, we derive a novel class of transformers, PID-controlled Transformer (PIDformer), aimed at improving robustness and mitigating the rank-collapse issue inherent in softmax transformers. We empirically evaluate the model for advantages and robustness against baseline transformers across various practical tasks, including object classification, image segmentation, and language modeling.

cs.AI↗

arXiv2023

Mitigating Over-smoothing in Transformers via Regularized Nonlocal Functionals

Transformers have achieved remarkable success in a wide range of natural language processing and computer vision applications. However, the representation capacity of a deep transformer model is degraded due to the over-smoothing issue in which the token representations become identical when the model's depth grows. In this work, we show that self-attention layers in transformers minimize a functional which promotes smoothness, thereby causing token uniformity. We then propose a novel regularizer that penalizes the norm of the difference between the smooth output tokens from self-attention and the input tokens to preserve the fidelity of the tokens. Minimizing the resulting regularized energy functional, we derive the Neural Transformer with a Regularized Nonlocal Functional (NeuTRENO), a novel class of transformer models that can mitigate the over-smoothing issue. We empirically demonstrate the advantages of NeuTRENO over the baseline transformers and state-of-the-art methods in reducing the over-smoothing of token representations on various practical tasks, including object classification, image segmentation, and language modeling.

cs.CL↗

arXiv2023

p-Laplacian Transformer

$p$-Laplacian regularization, rooted in graph and image signal processing, introduces a parameter $p$ to control the regularization effect on these data. Smaller values of $p$ promote sparsity and interpretability, while larger values encourage smoother solutions. In this paper, we first show that the self-attention mechanism obtains the minimal Laplacian regularization ($p=2$) and encourages the smoothness in the architecture. However, the smoothness is not suitable for the heterophilic structure of self-attention in transformers where attention weights between tokens that are in close proximity and non-close ones are assigned indistinguishably. From that insight, we then propose a novel class of transformers, namely the $p$-Laplacian Transformer (p-LaT), which leverages $p$-Laplacian regularization framework to harness the heterophilic features within self-attention layers. In particular, low $p$ values will effectively assign higher attention weights to tokens that are in close proximity to the current token being processed. We empirically demonstrate the advantages of p-LaT over the baseline transformers on a wide range of benchmark datasets.

cs.LG↗

arXiv2023

Characterizing Barriers and Technology Needs in the Kitchen for Blind and Low Vision People

Cooking is a vital yet challenging activity for people with visual impairments (PVI). It involves tasks that can be dangerous or difficult without vision, such as handling a knife or adding a suitable amount of salt. A better understanding of these challenges can inform the design of technologies that mitigate safety hazards and improve the quality of the lives of PVI. Furthermore, there is a need to understand the effects of different visual abilities, including low vision and blindness, and the role of rehabilitation training where PVI learn cooking skills and assistive technologies. In this paper, we aim to comprehensively characterize PVI's challenges, strategies, and needs in the kitchen from the perspectives of both PVI and rehabilitation professionals. Through a contextual inquiry study, we observed 10 PVI, including six low vision and four blind participants, when they cooked dishes of their choices in their own kitchens. We then interviewed six rehabilitation professionals to explore their training strategies and technology recommendations. Our findings revealed the differences between low vision and blind people during cooking as well as the gaps between training and reality. We suggest improvements for rehabilitation training and distill design considerations for future assistive technology in the kitchen.

cs.HC↗

arXiv2022

Transformer with Fourier Integral Attentions

Multi-head attention empowers the recent success of transformers, the state-of-the-art models that have achieved remarkable success in sequence modeling and beyond. These attention mechanisms compute the pairwise dot products between the queries and keys, which results from the use of unnormalized Gaussian kernels with the assumption that the queries follow a mixture of Gaussian distribution. There is no guarantee that this assumption is valid in practice. In response, we first interpret attention in transformers as a nonparametric kernel regression. We then propose the FourierFormer, a new class of transformers in which the dot-product kernels are replaced by the novel generalized Fourier integral kernels. Different from the dot-product kernels, where we need to choose a good covariance matrix to capture the dependency of the features of data, the generalized Fourier integral kernels can automatically capture such dependency and remove the need to tune the covariance matrix. We theoretically prove that our proposed Fourier integral kernels can efficiently approximate any key and query distributions. Compared to the conventional transformers with dot-product attention, FourierFormers attain better accuracy and reduce the redundancy between attention heads. We empirically corroborate the advantages of FourierFormers over the baseline transformers in a variety of practical applications including language modeling and image classification.

cs.LG↗

arXiv2021

Improving Transformers with Probabilistic Attention Keys

Multi-head attention is a driving force behind state-of-the-art transformers, which achieve remarkable performance across a variety of natural language processing (NLP) and computer vision tasks. It has been observed that for many applications, those attention heads learn redundant embedding, and most of them can be removed without degrading the performance of the model. Inspired by this observation, we propose Transformer with a Mixture of Gaussian Keys (Transformer-MGK), a novel transformer architecture that replaces redundant heads in transformers with a mixture of keys at each head. These mixtures of keys follow a Gaussian mixture model and allow each attention head to focus on different parts of the input sequence efficiently. Compared to its conventional transformer counterpart, Transformer-MGK accelerates training and inference, has fewer parameters, and requires fewer FLOPs to compute while achieving comparable or better accuracy across tasks. Transformer-MGK can also be easily extended to use with linear attention. We empirically demonstrate the advantage of Transformer-MGK in a range of practical applications, including language modeling and tasks that involve very long sequences. On the Wikitext-103 and Long Range Arena benchmark, Transformer-MGKs with 4 heads attain comparable or better performance to the baseline transformers with 8 heads.

cs.LG↗

arXiv2021

Active Learning in Incomplete Label Multiple Instance Multiple Label Learning

In multiple instance multiple label learning, each sample, a bag, consists of multiple instances. To alleviate labeling complexity, each sample is associated with a set of bag-level labels leaving instances within the bag unlabeled. This setting is more convenient and natural for representing complicated objects, which have multiple semantic meanings. Compared to single instance labeling, this approach allows for labeling larger datasets at an equivalent labeling cost. However, for sufficiently large datasets, labeling all bags may become prohibitively costly. Active learning uses an iterative labeling and retraining approach aiming to provide reasonable classification performance using a small number of labeled samples. To our knowledge, only a few works in the area of active learning in the MIML setting are available. These approaches can provide practical solutions to reduce labeling cost but their efficacy remains unclear. In this paper, we propose a novel bag-class pair based approach for active learning in the MIML setting. Due to the partial availability of bag-level labels, we focus on the incomplete-label MIML setting for the proposed active learning approach. Our approach is based on a discriminative graphical model with efficient and exact inference. For the query process, we adapt active learning criteria to the novel bag-class pair selection strategy. Additionally, we introduce an online stochastic gradient descent algorithm to provide an efficient model update after each query. Numerical experiments on benchmark datasets illustrate the robustness of the proposed approach.

cs.LG↗

arXiv2019

All One Needs to Know about Fog Computing and Related Edge Computing Paradigms: A Complete Survey

With the Internet of Things (IoT) becoming part of our daily life and our environment, we expect rapid growth in the number of connected devices. IoT is expected to connect billions of devices and humans to bring promising advantages for us. With this growth, fog computing, along with its related edge computing paradigms, such as multi-access edge computing (MEC) and cloudlet, are seen as promising solutions for handling the large volume of security-critical and time-sensitive data that is being produced by the IoT. In this paper, we first provide a tutorial on fog computing and its related computing paradigms, including their similarities and differences. Next, we provide a taxonomy of research topics in fog computing, and through a comprehensive survey, we summarize and categorize the efforts on fog computing and its related computing paradigms. Finally, we provide challenges and future directions for research in fog computing.

cs.NI↗

arXiv2018

TESS Discovery of a Transiting Super-Earth in the $π$ Mensae System

We report the detection of a transiting planet around $π$ Mensae (HD 39091), using data from the Transiting Exoplanet Survey Satellite (TESS). The solar-type host star is unusually bright (V=5.7) and was already known to host a Jovian planet on a highly eccentric, 5.7-year orbit. The newly discovered planet has a size of $2.04\pm 0.05$ $R_\oplus$ and an orbital period of 6.27 days. Radial-velocity data from the HARPS and AAT/UCLES archives also displays a 6.27-day periodicity, confirming the existence of the planet and leading to a mass determination of $4.82\pm 0.85$ $M_\oplus$. The star's proximity and brightness will facilitate further investigations, such as atmospheric spectroscopy, asteroseismology, the Rossiter--McLaughlin effect, astrometry, and direct imaging.

astro-ph.EP↗

arXiv2018

Expected Yields of Planet discoveries from the TESS primary and extended missions

We present a prediction of the transiting exoplanet yield of the TESS primary mission, in order to guide follow-up observations and science projects utilizing TESS discoveries. Our new simulations differ from previous work by using (1) an updated photometric noise model that accounts for the nominal pointing jitter estimated through simulation prior to launch, (2) improved stellar parameters based on Gaia mission Data Release 2, (3) improved empirically-based simulation of multi-planet systems, (4) a realistic method of selecting targets for 2-minute exposures, and (5) a more realistic geometric distortion model to determine the sky region that falls on TESS CCDs. We also present simulations of the planet yield for three suggested observing strategies of the TESS extended mission. We report ~$10^4$ planets to be discovered by the TESS primary mission, as well as an additional $\sim 2000$ planets for each year of the three extended mission scenarios we explored. We predict that in the primary mission, TESS will discover about 3500 planets with Neptune size and smaller, half of which will orbit stars with TESS magnitudes brighter than 12. Specifically, we proposed a new extended mission scenario that centers Camera 3 on the ecliptic pole (C3PO), which will yield more long period planets as well as moderately irradiated planets that orbit F, G, and K stars.

astro-ph.EP↗

arXiv2017

Explicit Reference Governor for the Constrained Control of Time-Delayed Linear Systems

This paper introduces an explicit reference governor approach for controlling time delay linear systems subject to state and input constraints. The proposed framework relies on suitable invariant sets that can be built using both Lyapunov-Razumikhin and Lyapunov-Krasovskii arguments. The proposed method is validated both numerically and experimentally using several alternative formulations.

eess.SY↗