arXiv Science⌕ Search

arXiv subjects

Andrew Wilson

Publications and source records attributed to Andrew Wilson.

11 recordsLinked to original sources
arXiv2026

From Gaze to Guidance: Interpreting and Adapting to Users' Cognitive Needs with Multimodal Gaze-Aware AI Assistants

Current LLM assistants are powerful at answering questions, but they have limited access to the behavioral context that reveals when and where a user is struggling. We present a gaze-grounded multimodal LLM assistant that uses egocentric video with gaze overlays to identify likely points of difficulty and target follow-up retrospective assistance. We instantiate this vision in a controlled study (n=36) comparing the gaze-aware AI assistant to a text-only LLM assistant. Compared to a conventional LLM assistant, the gaze-aware assistant was rated as significantly more accurate and personalized in its assessments of users' reading behavior and significantly improved people's ability to recall information. Users spoke significantly fewer words with the gaze-aware assistant, indicating more efficient interactions. Qualitative results underscored both perceived benefits in comprehension and challenges when interpretations of gaze behaviors were inaccurate. Our findings suggest that gaze-aware LLM assistants can reason about cognitive needs to improve cognitive outcomes of users.

cs.HC↗
arXiv2026

Reliable and Responsible Foundation Models: A Comprehensive Survey

Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), and Video Generative Models, have become essential tools with broad applications across various domains such as law, medicine, education, finance, science, and beyond. As these models see increasing real-world deployment, ensuring their reliability and responsibility has become critical for academia, industry, and government. This survey addresses the reliable and responsible development of foundation models. We explore critical issues, including bias and fairness, security and privacy, uncertainty, explainability, and distribution shift. Our research also covers model limitations, such as hallucinations, as well as methods like alignment and Artificial Intelligence-Generated Content (AIGC) detection. For each area, we review the current state of the field and outline concrete future research directions. Additionally, we discuss the intersections between these areas, highlighting their connections and shared challenges. We hope our survey fosters the development of foundation models that are not only powerful but also ethical, trustworthy, reliable, and socially responsible.

cs.LG↗
arXiv2023

An Application of the Causal Roadmap in Two Safety Monitoring Case Studies: Covariate-Adjustment and Outcome Prediction using Electronic Health Record Data

Real-world data, such as administrative claims and electronic health records, are increasingly used for safety monitoring and to help guide regulatory decision-making. In these settings, it is important to document analytic decisions transparently and objectively to ensure that analyses meet their intended goals. The Causal Roadmap is an established framework that can guide and document analytic decisions through each step of the analytic pipeline, which will help investigators generate high-quality real-world evidence. In this paper, we illustrate the utility of the Causal Roadmap using two case studies previously led by workgroups sponsored by the Sentinel Initiative -- a program for actively monitoring the safety of regulated medical products. Each case example focuses on different aspects of the analytic pipeline for drug safety monitoring. The first case study shows how the Causal Roadmap encourages transparency, reproducibility, and objective decision-making for causal analyses. The second case study highlights how this framework can guide analytic decisions beyond inference on causal parameters, improving outcome ascertainment in clinical phenotyping. These examples provide a structured framework for implementing the Causal Roadmap in safety surveillance and guide transparent, reproducible, and objective analysis.

stat.ME↗
arXiv2022

Uncertainty Calibration in Bayesian Neural Networks via Distance-Aware Priors

As we move away from the data, the predictive uncertainty should increase, since a great variety of explanations are consistent with the little available information. We introduce Distance-Aware Prior (DAP) calibration, a method to correct overconfidence of Bayesian deep learning models outside of the training domain. We define DAPs as prior distributions over the model parameters that depend on the inputs through a measure of their distance from the training set. DAP calibration is agnostic to the posterior inference method, and it can be performed as a post-processing step. We demonstrate its effectiveness against several baselines in a variety of classification and regression problems, including benchmarks designed to test the quality of predictive distributions away from the data.

stat.ML↗
arXiv2021

A non-parametric Bayesian approach for adjusting partial compliance in sequential decision making

Existing methods in estimating the mean outcome under a given dynamic treatment regime rely on intention-to-treat analyses which estimate the effect of following a certain dynamic treatment regime regardless of compliance behavior of patients. There are two major concerns with intention-to-treat analyses: (1) the estimated effects are often biased toward the null effect; (2) the results are not generalizable and reproducible due to the potential differential compliance behavior. These are particularly problematic in settings with high level of non-compliance such as substance use disorder treatments. Our work is motivated by the Adaptive Treatment for Alcohol and Cocaine Dependence study (ENGAGE), which is a multi-stage trial that aimed to construct optimal treatment strategies to engage patients in therapy. Due to the relatively low level of compliance in this trial, intention-to-treat analyses essentially estimate the effect of being randomized to a certain treatment sequence which is not of interest. We fill this important gap by defining the target parameter as the mean outcome under a dynamic treatment regime given potential compliance strata. We propose a flexible non-parametric Bayesian approach, which consists of a Gaussian copula model for the potential compliances, and a Dirichlet process mixture model for the potential outcomes. Our simulations highlight the need for and usefulness of this approach in practice and illustrate the robustness of our estimator in non-linear and non-Gaussian settings.

stat.ME↗
arXiv2015

Scalable Gaussian Processes for Characterizing Multidimensional Change Surfaces

We present a scalable Gaussian process model for identifying and characterizing smooth multidimensional changepoints, and automatically learning changes in expressive covariance structure. We use Random Kitchen Sink features to flexibly define a change surface in combination with expressive spectral mixture kernels to capture the complex statistical structure. Finally, through the use of novel methods for additive non-separable kernels, we can scale the model to large datasets. We demonstrate the model on numerical and real world data, including a large spatio-temporal disease dataset where we identify previously unknown heterogeneous changes in space and time.

stat.ML↗
arXiv2014

Digital frequency domain multiplexing readout electronics for the next generation of millimeter telescopes

Frequency domain multiplexing (fMux) is an established technique for the readout of transition-edge sensor (TES) bolometers in millimeter-wavelength astrophysical instrumentation. In fMux, the signals from multiple detectors are read out on a single pair of wires reducing the total cryogenic thermal loading as well as the cold component complexity and cost of a system. The current digital fMux system, in use by POLARBEAR, EBEX, and the South Pole Telescope, is limited to a multiplexing factor of 16 by the dynamic range of the Superconducting Quantum Interference Device pre-amplifier and the total system bandwidth. Increased multiplexing is key for the next generation of large format TES cameras, such as SPT-3G and POLARBEAR2, which plan to have on the of order 15,000 detectors. Here, we present the next generation fMux readout, focusing on the warm electronics. In this system, the multiplexing factor increases to 64 channels per module (2 wires) while maintaining low noise levels and detector stability. This is achieved by increasing the system bandwidth, reducing the dynamic range requirements though active feedback, and digital synthesis of voltage biases with a novel polyphase filter algorithm. In addition, a version of the new fMux readout includes features such as low power consumption and radiation-hard components making it viable for future space-based millimeter telescopes such as the LiteBIRD satellite.

astro-ph.IM↗
arXiv2011

A discrete random model describing bedrock profile abrasion

We use a simple, collision-based, discrete, random abrasion model to compute the profiles for the stoss faces in a bedrock abrasion process. The model is the discrete equivalent of the generalized version of a classical, collision based model of abrasion. Three control parameters (which describe the average size of the colliding objects, the expected direction of the impacts and the average volume removed from the body due to one collision) are sufficient for realistic predictions. Our computations show the robust emergence of steady state shapes, both the geometry and the time evolution of which shows good quantitative agreement with laboratory experiments.

physics.geo-ph↗
arXiv2010

On the Erdos-Straus conjecture

Paul Erdos conjectured that for every n in N, n>1, there exist a, b, c natural numbers, not necessarily distinct, so that 4/n=1/a+1/b+1/c (see \cite{rg}). In this paper we prove an extension of Mordell's theorem and formulate a conjecture which is stronger than Erdos' conjecture.

math.NT↗
arXiv1998

Ultraviolet-Optical observations of the Seyfert 2 Galaxies NGC 7130, NGC 5135 and IC 3639: Implications for the Starburst-AGN Connection

We present and discuss HST (WFPC2 and FOC) images and UV GHRS spectra plus ground-based near UV through to near IR spectra of three Seyfert 2 nuclei (NGC 7130, NGC 5135 and IC 3639). These galaxies, together to Mrk 477, were selected from a bigger sample that comprises the 20 brightest Seyfert 2 nuclei, with the goal to study the origin of the UV-optical-near IR featureless continuum in Seyfert 2 nuclei. These four galaxies have bolometric luminosities, as computed with the four IRAS bands, of 10^11 Lsol. They are close enough to be resolved with HST the nuclear zone. This makes these Seyfert 2 galaxies benchmarks to study the Starburst-AGN connection in more distant galaxies. The data provide direct evidence of the existence of a central nuclear starburst that dominates the UV light, and that seem to be responsible for the origin of the so called featureless continuum. These starbursts are dusty and compact. They have sizes (from less than 100 pc to a few hundred pc) much smaller and closer to the nucleus than that seen in the prototype Seyfert 2 galaxy NGC 1068. The bolometric luminosity of these starbursts is similar to the estimated bolometric luminosities of their obscured Seyfert 1 nuclei, and thus they contribute in the same amount to the overall energetics of these galaxies.

astro-ph↗