arXiv subjects

Young Lee

Publications and source records attributed to Young Lee.

At least 19 recordsLinked to original sources

arXiv2026

Longitudinal Outcomes Truncated by Death: Causal Estimands and Bayesian Estimators

In randomized controlled trials with longitudinal outcomes, death before the end of follow-up poses a fundamental challenge: after death, the outcome is no longer a real-valued measurement. This complicates the definition and interpretation of causal estimands, particularly when treatment may affect both survival and longitudinal outcomes. We review existing estimands for longitudinal outcomes truncated by death and clarify the assumptions required for their identification and estimation. We show that these estimands fall into two broad classes, distinguished by whether they require additional assumptions to compare longitudinal outcomes beyond death. Such assumptions may be inappropriate in chronic diseases, either because i) death and longitudinal outcomes are driven by the same underlying biological process or ii) the relative desirability of survival with poor function versus death may depend on individual preferences. We compare the behavior of the estimands in a simulation study using Bayesian estimators and illustrate their use with data from a randomized controlled trial in amyotrophic lateral sclerosis. We argue that, in the presence of death truncation, pairing the survivor average causal effect with the restricted mean survival time estimand provides an interpretable characterization of treatment effects on longitudinal and survival outcomes.

stat.ME↗

arXiv2026

Symbolic Execution Meets Multi-LLM Orchestration: Detecting Memory Vulnerabilities in Incomplete Rust CVE Snippets

This paper presents a system combining symbolic execution (KLEE) with a 4-agent multi-LLM architecture for detecting memory vulnerabilities in Rust unsafe code. A central challenge we address is the incomplete-code problem: CVE database entries provide only isolated code snippets that lack struct definitions, imports, and Cargo manifests, causing all existing formal verification tools to fail at compilation with zero output. Our system resolves this through four specialized agents -- an Oracle/Validator for strategic planning, a Safety Checker for vulnerability analysis, a Code Specialist for FFI wrapper generation, and a Fast Filter for execution optimization -- that collaboratively synthesize KLEE-compatible harnesses from otherwise uncompilable fragments. KLEE's output is then ingested by graph_klee.py, which constructs a Graph Database linking CVE files, CWE categories, error types, and symbolic execution paths as typed nodes and labelled edges, enabling structured cross-CVE vulnerability queries. We evaluated our system on 31 real-world Rust CVEs spanning 11 CWE categories, achieving 90.3% wrapper compilation success where all state-of-the-art formal verification tools achieve 0%. Our system detected 1,206 critical errors across 26 files (83.9% detection rate), compared to 14 warnings across 11 files for Clippy (35.5%) and generic labels for Miri. The 4-agent architecture reduced wrapper compilation failures from 42% (single-agent baseline) to 9.7% and increased detected errors from 487 to 1,206, confirming that role specialization and structured context passing produce measurably better results than a single general-purpose model. Our replication package is publicly available at https://github.com/Zeyad-Ab/Symbolic-Execution-with-Multi-LLM-Architecture-for-Rust-Security

cs.CR↗

arXiv2026

MedProbCLIP: Probabilistic Adaptation of Vision-Language Foundation Model for Reliable Radiograph-Report Retrieval

Vision-language foundation models have emerged as powerful general-purpose representation learners with strong potential for multimodal understanding, but their deterministic embeddings often fail to provide the reliability required for high-stakes biomedical applications. This work introduces MedProbCLIP, a probabilistic vision-language learning framework for chest X-ray and radiology report representation learning and bidirectional retrieval. MedProbCLIP models image and text representations as Gaussian embeddings through a probabilistic contrastive objective that explicitly captures uncertainty and many-to-many correspondences between radiographs and clinical narratives. A variational information bottleneck mitigates overconfident predictions, while MedProbCLIP employs multi-view radiograph encoding and multi-section report encoding during training to provide fine-grained supervision for clinically aligned correspondence, yet requires only a single radiograph and a single report at inference. Evaluated on the MIMIC-CXR dataset, MedProbCLIP outperforms deterministic and probabilistic baselines, including CLIP, CXR-CLIP, and PCME++, in both retrieval and zero-shot classification. Beyond accuracy, MedProbCLIP demonstrates superior calibration, risk-coverage behavior, selective retrieval reliability, and robustness to clinically relevant corruptions, underscoring the value of probabilistic vision-language modeling for improving the trustworthiness and safety of radiology image-text retrieval systems.

cs.CV↗

arXiv2025

Prompting the Priorities: A First Look at Evaluating LLMs for Vulnerability Triage and Prioritization

Security analysts face increasing pressure to triage large and complex vulnerability backlogs. Large Language Models (LLMs) offer a potential aid by automating parts of the interpretation process. We evaluate four models (ChatGPT, Claude, Gemini, and DeepSeek) across twelve prompting techniques to interpret semi-structured and unstructured vulnerability information. As a concrete use case, we test each model's ability to predict decision points in the Stakeholder-Specific Vulnerability Categorization (SSVC) framework: Exploitation, Automatable, Technical Impact, and Mission and Wellbeing. Using 384 real-world vulnerabilities from the VulZoo dataset, we issued more than 165,000 queries to assess performance under prompting styles including one-shot, few-shot, and chain-of-thought. We report F1 scores for each SSVC decision point and Cohen's kappa (weighted and unweighted) for the final SSVC decision outcomes. Gemini consistently ranked highest, leading on three of four decision points and yielding the most correct recommendations. Prompting with exemplars generally improved accuracy, although all models struggled on some decision points. Only DeepSeek achieved fair agreement under weighted metrics, and all models tended to over-predict risk. Overall, current LLMs do not replace expert judgment. However, specific LLM and prompt combinations show moderate effectiveness for targeted SSVC decisions. When applied with care, LLMs can support vulnerability prioritization workflows and help security teams respond more efficiently to emerging threats.

cs.CR↗

arXiv2025

Causal Policy Learning in Reinforcement Learning: Backdoor-Adjusted Soft Actor-Critic

Hidden confounders that influence both states and actions can bias policy learning in reinforcement learning (RL), leading to suboptimal or non-generalizable behavior. Most RL algorithms ignore this issue, learning policies from observational trajectories based solely on statistical associations rather than causal effects. We propose DoSAC (Do-Calculus Soft Actor-Critic with Backdoor Adjustment), a principled extension of the SAC algorithm that corrects for hidden confounding via causal intervention estimation. DoSAC estimates the interventional policy $\pi(a | \mathrm{do}(s))$ using the backdoor criterion, without requiring access to true confounders or causal labels. To achieve this, we introduce a learnable Backdoor Reconstructor that infers pseudo-past variables (previous state and action) from the current state to enable backdoor adjustment from observational data. This module is integrated into a soft actor-critic framework to compute both the interventional policy and its entropy. Empirical results on continuous control benchmarks show that DoSAC outperforms baselines under confounded settings, with improved robustness, generalization, and policy reliability.

cs.LG↗

arXiv2025

Benchmarking Robustness of Contrastive Learning Models for Medical Image-Report Retrieval

Medical images and reports offer invaluable insights into patient health. The heterogeneity and complexity of these data hinder effective analysis. To bridge this gap, we investigate contrastive learning models for cross-domain retrieval, which associates medical images with their corresponding clinical reports. This study benchmarks the robustness of four state-of-the-art contrastive learning models: CLIP, CXR-RePaiR, MedCLIP, and CXR-CLIP. We introduce an occlusion retrieval task to evaluate model performance under varying levels of image corruption. Our findings reveal that all evaluated models are highly sensitive to out-of-distribution data, as evidenced by the proportional decrease in performance with increasing occlusion levels. While MedCLIP exhibits slightly more robustness, its overall performance remains significantly behind CXR-CLIP and CXR-RePaiR. CLIP, trained on a general-purpose dataset, struggles with medical image-report retrieval, highlighting the importance of domain-specific training data. The evaluation of this work suggests that more effort needs to be spent on improving the robustness of these models. By addressing these limitations, we can develop more reliable cross-domain retrieval models for medical applications.

cs.CV↗

arXiv2024

Hawkes Models And Their Applications

The Hawkes process is a model for counting the number of arrivals to a system which exhibits the self-exciting property - that one arrival creates a heightened chance of further arrivals in the near future. The model, and its generalizations, have been applied in a plethora of disparate domains, though two particularly developed applications are in seismology and in finance. As the original model is elegantly simple, generalizations have been proposed which: track marks for each arrival, are multivariate, have a spatial component, are driven by renewal processes, treat time as discrete, and so on. This paper creates a cohesive review of the traditional Hawkes model and the modern generalizations, providing details on their construction, simulation algorithms, and giving key references to the appropriate literature for a detailed treatment.

stat.ME↗

arXiv2023

Translating Natural Language Queries to SQL Using the T5 Model

This paper presents the development process of a natural language to SQL model using the T5 model as the basis. The models, developed in August 2022 for an online transaction processing system and a data warehouse, have a 73\% and 84\% exact match accuracy respectively. These models, in conjunction with other work completed in the research project, were implemented for several companies and used successfully on a daily basis. The approach used in the model development could be implemented in a similar fashion for other database environments and with a more powerful pre-trained language model.

cs.DB↗

arXiv2023

An Adaptive Kernel Approach to Federated Learning of Heterogeneous Causal Effects

We propose a new causal inference framework to learn causal effects from multiple, decentralized data sources in a federated setting. We introduce an adaptive transfer algorithm that learns the similarities among the data sources by utilizing Random Fourier Features to disentangle the loss function into multiple components, each of which is associated with a data source. The data sources may have different distributions; the causal effects are independently and systematically incorporated. The proposed method estimates the similarities among the sources through transfer coefficients, and hence requiring no prior information about the similarity measures. The heterogeneous causal effects can be estimated with no sharing of the raw training data among the sources, thus minimizing the risk of privacy leak. We also provide minimax lower bounds to assess the quality of the parameters learned from the disparate sources. The proposed method is empirically shown to outperform the baselines on decentralized data sources with dissimilar distributions.

cs.LG↗

arXiv2022

Enhanced superconductivity by near-neighbor attraction in the doped Hubbard model

Recent experiment has unveiled an anomalously strong electron-electron attraction in one-dimensional copper-oxide chain Ba$_{2-x}$Sr$_x$CuO$_{3+\delta}$. While the near-neighbor electron attraction $V$ in the one-dimensional extended Hubbard chain has been examined recently, its effect in the Hubbard model beyond the one-dimensional chain remains unclear. We report a density-matrix renormalization group study of the extended Hubbard model on long four-leg cylinders on the square lattice. We find that the near-neighbor electron attraction $V$ can notably enhance the long-distance superconducting correlations while simultaneously suppressing the charge-density-wave correlations. Specifically, for a modestly strong electron attraction, the superconducting correlations become dominant over the CDW correlations with a Luttinger exponent $K_{sc}\sim 1$ and strong divergent superconducting susceptibility. Our results provide a promising way to realize long-range superconductivity in the doped Hubbard model in two dimensions. The relevance of our numerical results to cuprate materials is also discussed.

cond-mat.str-el↗

arXiv2021

Exact simulation of extrinsic stress-release processes

We present a new and straightforward algorithm that simulates exact sample paths for a generalized stress-release process. The computation of the exact law of the joint interarrival times is detailed and used to derive this algorithm. Furthermore, the martingale generator of the process is derived and induces theoretical moments which generalize some results of Borovkov & Vere-Jones (2000) and are used to demonstrate the validity of our simulation algorithm.

stat.CO↗

arXiv2021

Federated Estimation of Causal Effects from Observational Data

Many modern applications collect data that comes in federated spirit, with data kept locally and undisclosed. Till date, most insight into the causal inference requires data to be stored in a central repository. We present a novel framework for causal inference with federated data sources. We assess and integrate local causal effects from different private data sources without centralizing them. Then, the treatment effects on subjects from observational data using a non-parametric reformulation of the classical potential outcomes framework is estimated. We model the potential outcomes as a random function distributed by Gaussian processes, whose defining parameters can be efficiently learned from multiple data sources, respecting privacy constraints. We demonstrate the promise and efficiency of the proposed approach through a set of simulated and real-world benchmark examples.

stat.ME↗

arXiv2020

Bayesian causal inference for count potential outcomes

The literature for count modeling provides useful tools to conduct causal inference when outcomes take non-negative integer values. Applied to the potential outcomes framework, we link the Bayesian causal inference literature to statistical models for count data. We discuss the general architectural considerations for constructing the predictive posterior of the missing potential outcomes. Special considerations for estimating average treatment effects are discussed, some generalizing certain relationships and some not yet encountered in the causal inference literature.

stat.ME↗

arXiv2019

Generative Parameter Sampler For Scalable Uncertainty Quantification

Uncertainty quantification has been a core of the statistical machine learning, but its computational bottleneck has been a serious challenge for both Bayesians and frequentists. We propose a model-based framework in quantifying uncertainty, called predictive-matching Generative Parameter Sampler (GPS). This procedure considers an Uncertainty Quantification (UQ) distribution on the targeted parameter, which matches the corresponding predictive distribution to the observed data. This framework adopts a hierarchical modeling perspective such that each observation is modeled by an individual parameter. This individual parameterization permits the resulting inference to be computationally scalable and robust to outliers. Our approach is illustrated for linear models, Poisson processes, and deep neural networks for classification. The results show that the GPS is successful in providing uncertainty quantification as well as additional flexibility beyond what is allowed by classical statistical procedures under the postulated statistical models.

cs.LG↗

arXiv2018

A magneto-optical trap created by the 2nd-order external cavity diode lasers

In this article, we report on a magneto-optical trap (MOT) created by the 2nd-order external cavity diode lasers (ECDLs). The lasers were characterized. We have observed the non-continuous changes of the wavelength as a function of the laser diode current. This study is beneficial for achieving tunable atom-atom interactions, quantum tunneling, precision measurement, ultracold plasma, as well as quantum computing.

physics.ins-det↗

arXiv2018

Unexpected Zero Bias Conductance Peak on the Topological Semimetal Sb(111) with a Broken Bilayer

The long-sought Majorana fermion is expected to manifest in a topological-superconductor heterostructure as a zero bias conductance peak (ZBCP). As one promising platform for such heterostructures, we investigate the cleaved surface of the topological semimetal Sb(111) using scanning tunneling microscopy and spectroscopy. Remarkably, we find a robust ZBCP on some terraces of the cleaved surface, although no superconductor is present. Using quasiparticle interference imaging, Landau level spectroscopy and density functional theory, we show that the ZBCP originates from a van Hove singularity pushed up to the Fermi level by a sub-surface stacking fault. Amidst the sprint to stake claims on new Majorana fermion systems, our finding highlights the importance of using a local probe together with detailed modeling to check thoroughly for crystal imperfections that may give rise to a trivial ZBCP unrelated to Majorana physics.

cond-mat.mes-hall↗

arXiv2018

Simultaneous optimization of isocenter locations and sector duration in radiosurgery

Stereotactic radiosurgery is an effective technique to treat brain tumors for which several inverse planning methods may be appropriate. We propose an integer programming model to simultaneous sector duration and isocenter optimization (SDIO) problem for Leksell Gamma Knife{\textregistered} Icon{\texttrademark} (Elekta, Stockholm, Sweden) to tractably incorporate treatment time. We devise a Benders decomposition scheme to solve the SDIO problem to optimality. The performances of our approaches are assessed using anonymized data from eight previously treated cases, and obtained treatment plans are compared against each other and against the clinical plans. The plans generated by our SDIO model all meet or exceed clinical guidelines while demonstrating high conformity.

physics.med-ph↗

arXiv2018

Simulation and Calibration of a Fully Bayesian Marked Multidimensional Hawkes Process with Dissimilar Decays

We propose a simulation method for multidimensional Hawkes processes based on superposition theory of point processes. This formulation allows us to design efficient simulations for Hawkes processes with differing exponentially decaying intensities. We demonstrate that inter-arrival times can be decomposed into simpler auxiliary variables that can be sampled directly, giving exact simulation with no approximation. We establish that the auxiliary variables provides information on the parent process for each event time. The algorithm correctness is shown by verifying the simulated intensities with their theoretical moments. A modular inference procedure consisting of Gibbs samplers through the auxiliary variable augmentation and adaptive rejection sampling is presented. Finally, we compare our proposed simulation method against existing methods, and find significant improvement in terms of algorithm speed. Our inference algorithm is used to discover the strengths of mutually excitations in real dark networks.

stat.ML↗