arXiv subjects

Junwei Ma

Publications and source records attributed to Junwei Ma.

At least 19 recordsLinked to original sources

arXiv2026

Understanding the Surprising Generalization Properties of Tabular Foundation Models

Tabular Foundation Models (TFMs) increasingly rely on in-context learning, where a model receives labelled examples at inference time and predicts labels for new inputs without updating its weights. Existing TFMs are typically trained on either massive synthetic corpora or very large collections of real datasets. In contrast, we show that surprisingly strong transfer can emerge from self-supervised pre-training on just a single real table. In this setting, we also find that tables tend to be either broadly useful or broadly poor regardless of downstream prediction task, and that the strongest predictor of usefulness is the number of features rather than the number of instances. This leads to a task-centric interpretation of tabular pre-training: the number and the quality of tasks are essential for the pre-training of TFMs. We show that the same task-centric perspective can help corpus design at scale: fine-grained column-level pre-processing consistently improves downstream performance, while no improvements are observed when we filter or deduplicate at the dataset level. Finally, we offer a new perspective for how TFMs generalize: we believe that tabular in-context generalization is largely retrieval-based, and good models are those that learn to identify relevant examples in the provided context and aggregate them well. The mechanics of TFMs have been relatively understudied; our task-centric, retrieval-based perspective offers a new framework to guide future model and corpus design.

cs.LG↗

arXiv2026

Damage-TriageFormer: A Foundation-Model Framework for Typology-Based Building Damage Assessment from Mono-Temporal Imagery

Decision-relevant building damage assessment is critical for prioritizing resources and recovery after a disaster, yet most automated methods either flatten damage into a single severity scale (no damage, minor, major, destroyed) or require paired pre- and post-event imagery that is often unavailable for emerging hazards. This paper presents Damage-TriageFormer, a single-image, post-event, footprint-conditioned model that produces a damage typology rather than a severity scale. We contribute: (1) DamageTriage-Bench, a new benchmark built from NOAA Emergency Response Imagery across Hurricane Michael (2018), Hurricane Helene (2024), and the 2025 Los Angeles wildfire complex, with five typology classes that distinguish roof damage from structural damage and, within each, partial from total extent; and (2) Damage-TriageFormer, which extends a DINOv3 ViT-L backbone with a Simple Feature Pyramid for higher-resolution instance pooling, a two-stage gated damage head, and an auxiliary severity-regression objective. Our model achieves macro F1 of 0.624 on validation and 0.619 on a held-out stratified test set, performing strongest where operational triage needs it most, with per-class F1 of 0.91 and 0.84 on undamaged buildings and total structural collapse, respectively. While the rare Total Roof Damage class remains difficult due to its limited examples and an inherently ambiguous label boundary, our results show that single-image post-event imagery can support actionable building damage typing, enabling targeted emergency response and resource allocation without a pre-event reference.

cs.CV↗

arXiv2026

Symmetry Regularization of 1D Generalized Coulomb Problems

For the 1D generalized Coulomb problems--a family that includes the quantizations of the 1D generalized Kepler problems of Ma-Meng-Xiao \cite{MMX2025}--we construct two explicit unitary intertwiners $\hat\iota_{\pm}$, the quantum analogs of the classical symmetry regularization maps $\iota_{\pm}$ of \cite{MMX2025}, that unitarily identify each $H_{\kappa}$-energy-definite portion of the Hilbert space $L^{2}(\mathbb{R}_{>0},\mathrm{d}q)$ with a unitary lowest-weight representation of $\widetilde{\mathrm{SL}}(2,\mathbb{R})$.

math-ph↗

arXiv2026

Data-efficient flood depth prediction through domain-aware coreset selection and tabular foundation models

Near-real-time flood depth prediction demands surrogate models that are accurate, fast, and transferable across watersheds. Supervised surrogates can match physics-based simulators in accuracy but need millions of training rows per watershed and cannot extrapolate beyond their original mesh. We propose a domain-aware coreset construction pipeline that conditions a tabular foundation model at inference time. The pipeline stratifies storms by return period and most-affected watershed, then samples hexagons with a target-aware spatial selector. With 0.7% of the per-watershed training pool, the model attains a mean $R^2$ of 0.663 across nine Houston-area watersheds, within 98.5% of the supervised reference ($R^2$ = 0.673). It transfers to held-out watersheds without task-specific retraining, staying ahead of a coreset-trained supervised baseline. On real storms it exceeds the supervised reference on a far out-of-distribution case and trails it on a mostly in-distribution one. Domain-aware coreset construction lets tabular foundation models deliver data-efficient, watershed-transferable flood predictions without per-watershed training.

cs.LG↗

arXiv2026

DisastRAG: A Multi-Source Disaster Information Integration and Access System Based on Retrieval-Augmented Large Language Models

Effective disaster management requires rapid access to information distributed across structured operational records, unstructured institutional documents, and dynamic external sources. However, most existing disaster information systems and retrieval-augmented generation frameworks remain organized around a single access pathway, limiting their ability to support heterogeneous, time-sensitive, and context-dependent information needs. This study presents DisastRAG, a disaster-aware information integration and access system that combines large language models with retrieval-augmented access to structured, unstructured, and contextual disaster information. The framework is built around a multi-path architecture that supports document retrieval over a curated hazard corpus, structured access over relational disaster records, and external web fallback for out-of-corpus requests, while also incorporating query understanding, strategy routing, response generation, and contextual memory within a unified system. We evaluated the document retrieval performance using four open-source large language models across multiple retrieval configurations on multiple-choice and open-ended disaster information tasks. Retrieval augmentation consistently improves performance over no-retrieval baselines, yielding multiple-choice gains of 12-23 percentage points and open-ended keypoint coverage gains of up to 10.5 percentage points. Results show that larger candidate pools are most helpful for weaker models, while stronger models are more sensitive to retrieval noise. Hybrid retrieval performs best for open-ended coverage, whereas vector retrieval and shallower reranking more often favor closed-form factual selection. Case studies further show that structured access and web fallback extend the framework beyond document-only RAG.

cs.IR↗

arXiv2026

DisastQA: A Comprehensive Benchmark for Evaluating Question Answering in Disaster Management

Accurate question answering (QA) in disaster management requires reasoning over uncertain and conflicting information, a setting poorly captured by existing benchmarks built on clean evidence. We introduce DisastQA, a large-scale benchmark of 3,000 rigorously verified questions (2,000 multiple-choice and 1,000 open-ended) spanning eight disaster types. The benchmark is constructed via a human-LLM collaboration pipeline with stratified sampling to ensure balanced coverage. Models are evaluated under varying evidence conditions, from closed-book to noisy evidence integration, enabling separation of internal knowledge from reasoning under imperfect information. For open-ended QA, we propose a human-verified keypoint-based evaluation protocol emphasizing factual completeness over verbosity. Experiments with 20 models reveal substantial divergences from general-purpose leaderboards such as MMLU-Pro. While recent open-weight models approach proprietary systems in clean settings, performance degrades sharply under realistic noise, exposing critical reliability gaps for disaster response. All code, data, and evaluation resources are available at https://github.com/TamuChen18/DisastQA_open.

cs.CL↗

arXiv2025

A Parametric Framework for Anticipatory Flashflood Warning: Integrating Landscape Vulnerability with Precipitation Forecasts

Flash flood warnings are largely reactive, providing limited advance notice for evacuation planning and resource prepositioning. This study presents and validates an anticipatory, parametric framework that converts landscape vulnerability and precipitation into transparent, zone-aware threat levels at neighborhood scales. We first derive an inherent hazard likelihood (IHL) surface using pluvial flood depth, height above nearest drainage, and distance to streams. Next, we compute a hazard severity index (HSI) by normalizing 24-hour rainfall against local Atlas-14 100-year, 24-hour depths. We then integrate IHL and HSI within a localized threat severity (LTS) matrix using 20 class-specific triggers, requiring lower exceedance in high-risk terrain and higher exceedance in uplands. Applied to two Texas flood events, the LTS exhibits statistically significant spatial association with independent crowdsourced impact proxies, capturing observed disruption hotspots. The framework is computationally lightweight, scalable, and extends actionable situational awareness into a 48-72 hour anticipatory window, supporting pre-event decision-making by emergency managers.

cs.CE↗

arXiv2025

Quantifying Functional Criticality of Lifelines Through Mobility-Derived Population-Facility Dependence for Human-Centered Resilience

Lifeline infrastructure underpins the continuity of daily life, yet conventional criticality assessments remain largely asset-centric, inferring importance from physical capacity or network topology rather than actual behavioral reliance. This disconnect frequently obscures the true societal cost of disruption, particularly in underserved communities where residents lack service alternatives. This study bridges the gap between engineering risk analysis and human mobility analysis by introducing functional criticality, a human-centered metric that quantifies the behavioral indispensability of specific facilities based on large-scale visitation patterns. Leveraging 1.02 million anonymized mobility records for Harris County, Texas, we operationalize lifeline criticality as a function of visitation intensity, catchment breadth, and origin-specific substitutability. Results reveal that dependence is highly concentrated: a small subset of super-critical facilities (2.8% of grocery stores and 14.8% of hospitals) supports a disproportionate share of routine access. By coupling these behavioral scores with probabilistic flood hazard models for 2020 and 2060, we demonstrate a pronounced risk-multiplier effect. While physical flood depths increase only moderately under future climate scenarios, the population-weighted vulnerability of access networks surges by 67.6%. This sharp divergence establishes that future hazards will disproportionately intersect with the specific nodes communities rely on most. The proposed framework advances resilience assessment by embedding human behavior directly into the definition of infrastructure importance, providing a scalable foundation for equitable, adaptive, and human-centered resilience planning.

cs.CY↗

arXiv2025

Decoupling Urban Food Accessibility Resilience during Disasters through Time-Series Analysis of Human Mobility and Power Outages

Disaster-induced power outages create cascading disruptions across urban lifelines, yet the timed coupling between grid failure and essential service access remains poorly quantified. Focusing on Hurricane Beryl in Houston (2024), this study integrates approximately 173000 15-minute outage records with over 1.25 million visits to 3187 food facilities to quantify how infrastructure performance and human access co-evolve. We construct daily indices for outage characteristics (intensity, duration) and food access metrics (redundancy, frequency, proximity), estimate cross-system lags through lagged correlations over zero to seven days, and identify recovery patterns using DTW k-means clustering. Overlaying these clusters yields compound power-access typologies and enables facility-level criticality screening. The analysis reveals a consistent two-day lag: food access reaches its nadir on July 8 at landfall while outage severity peaks around July 10, with negative correlations strongest at a two-day lag and losing significance by day four. We identify four compound typologies from high/low outage crossed with high/low access disruption levels. Road network sparsity, more than income, determines the depth and persistence of access loss. Through this analysis, we enumerate 294 critical food facilities in the study area requiring targeted continuity measures including backup power, microgrids, and feeder prioritization. The novelty lies in measuring interdependency at daily operational resolution while bridging scales from communities to individual facilities, converting dynamic coupling patterns into actionable interventions for phase-sensitive restoration and equity-aware preparedness. The framework is transferable to other lifelines and hazards, offering a generalizable template for diagnosing and mitigating cascading effects on community access during disaster recovery.

stat.AP↗

arXiv2025

Generalization Can Emerge in Tabular Foundation Models From a Single Table

Deep tabular modelling increasingly relies on in-context learning where, during inference, a model receives a set of $(x,y)$ pairs as context and predicts labels for new inputs without weight updates. We challenge the prevailing view that broad generalization here requires pre-training on large synthetic corpora (e.g., TabPFN priors) or a large collection of real data (e.g., TabDPT training datasets), discovering that a relatively small amount of data suffices for generalization. We find that simple self-supervised pre-training on just a \emph{single} real table can produce surprisingly strong transfer across heterogeneous benchmarks. By systematically pre-training and evaluating on many diverse datasets, we analyze what aspects of the data are most important for building a Tabular Foundation Model (TFM) generalizing across domains. We then connect this to the pre-training procedure shared by most TFMs and show that the number and quality of \emph{tasks} one can construct from a dataset is key to downstream performance.

cs.LG↗

arXiv2025

Disaster Management in the Era of Agentic AI Systems: A Vision for Collective Human-Machine Intelligence for Augmented Resilience

The escalating frequency and severity of disasters routinely overwhelm traditional response capabilities, exposing critical vulnerability in disaster management. Current practices are hindered by fragmented data streams, siloed technologies, resource constraints, and the erosion of institutional memory, which collectively impede timely and effective decision making. This study introduces Disaster Copilot, a vision for a multi-agent artificial intelligence system designed to overcome these systemic challenges by unifying specialized AI tools within a collaborative framework. The proposed architecture utilizes a central orchestrator to coordinate diverse sub-agents, each specializing in critical domains such as predictive risk analytics, situational awareness, and impact assessment. By integrating multi-modal data, the system delivers a holistic, real-time operational picture and serve as the essential AI backbone required to advance Disaster Digital Twins from passive models to active, intelligent environments. Furthermore, it ensures functionality in resource-limited environments through on-device orchestration and incorporates mechanisms to capture institutional knowledge, mitigating the impact of staff turnover. We detail the system architecture and propose a three-phased roadmap emphasizing the parallel growth of technology, organizational capacity, and human-AI teaming. Disaster Copilot offers a transformative vision, fostering collective human-machine intelligence to build more adaptive, data-driven and resilient communities.

cs.MA↗

arXiv2025

Quantifying the Social Costs of Power Outages and Restoration Disparities Across Four U.S. Hurricanes

The multifaceted nature of disaster impact shows that densely populated areas contribute more to aggregate burden, while sparsely populated but heavily affected regions suffer disproportionately at the individual level. This study introduces a framework for quantifying the societal impacts of power outages by translating customer weighted outage exposure into deprivation measures, integrating welfare metrics with three recovery indicators, average outage days per customer, restoration duration, and relative restoration rate, computed from sequential EAGLE I observations and linked to Zip Code Tabulation Area demographics. Applied to four United States hurricanes, Beryl 2024 Texas, Helene 2024 Florida, Milton 2024 Florida, and Ida 2021 Louisiana, this standardized pipeline provides the first cross event, fine scale evaluation of outage impacts and their drivers. Results demonstrate regressive patterns with greater burdens in lower income areas, mechanistic analysis shows deprivation increases with longer restoration durations and decreases with faster restoration rates, explainable modeling identifies restoration duration as the dominant driver, and clustering reveals distinct recovery typologies not captured by conventional reliability metrics. This framework delivers a transferable method for assessing outage impacts and equity, comparative cross event evidence linking restoration dynamics to social outcomes, and actionable spatial analyses that support equity informed restoration planning and resilience investment.

physics.soc-ph↗

arXiv2025

CausalPFN: Amortized Causal Effect Estimation via In-Context Learning

Causal effect estimation from observational data is fundamental across various applications. However, selecting an appropriate estimator from dozens of specialized methods demands substantial manual effort and domain expertise. We present CausalPFN, a single transformer that amortizes this workflow: trained once on a large library of simulated data-generating processes that satisfy ignorability, it infers causal effects for new observational datasets out of the box. CausalPFN combines ideas from Bayesian causal inference with the large-scale training protocol of prior-fitted networks (PFNs), learning to map raw observations directly to causal effects without any task-specific adjustment. Our approach achieves superior average performance on heterogeneous and average treatment effect estimation benchmarks (IHDP, Lalonde, ACIC). Moreover, it shows competitive performance for real-world policy making on uplift modeling tasks. CausalPFN provides calibrated uncertainty estimates to support reliable decision-making based on Bayesian principles. This ready-to-use model requires no further training or tuning and takes a step toward automated causal inference (https://github.com/vdblm/CausalPFN/).

cs.LG↗

arXiv2025

Anatomy of a Historic Blackout: Decoding Spatiotemporal Dynamics of Power Outages and Disparities During Hurricane Beryl

This study investigates the spatial patterns and temporal variations in outage duration, intensity, and restoration/recovery following the 2024 Hurricane Beryl in Houston, Texas. This historic blackout caused widespread power disruptions across the Houston metropolitan area, leaving more than 2 million customers without power over several days, resulting in more than 143 million total customer-out hours.The findings reveal that areas with higher population density and proximity to the hurricane's path experienced more severe initial impacts. Regions with higher median income showed faster recovery, while lower-income areas exhibited prolonged restoration periods, even with favorable infrastructural conditions, suggesting disparities in restoration speed. The study also highlights how urban development features, such as road density and land elevation, explain spatial disparities in power outage impacts and recovery. This research advances the understanding of power outage dynamics in large metropolitan regions through four key contributions: (1) empirical characterization of outages from a historic hurricane, highlighting infrastructure vulnerabilities in a high-density urban context; (2) comprehensive analysis using multiple metrics to capture spatiotemporal dynamics of outages and restoration; (3) leveraging of high-resolution outage data at fine geographic scales and frequent intervals to quantify and reveal previously masked spatial disparities; and (4) systematic examination of socioeconomic, urban development, and environmental factors in shaping disparities in outage impacts and recovery timelines. These findings provide infrastructure managers, operators, utilities, and decision-makers with crucial empirical insights to quantify power outage impacts, justify resilience investments, and address vulnerability and equity issues in the power infrastructure during hazard events.

cs.CE↗

arXiv2024

AI Meets Natural Hazard Risk: A Nationwide Vulnerability Assessment of Data Centers to Natural Hazards and Power Outages

Our society is on the verge of a revolution powered by Artificial Intelligence (AI) technologies. With increasing advancements in AI, there is a growing expansion in data centers (DCs) serving as critical infrastructure for this new wave of technologies. This technological wave is also on a collision course with exacerbating climate hazards which raises the need for evaluating the vulnerability of DCs to various hazards. Hence, the objective of this research is to conduct a nationwide vulnerability assessment of (DCs) in the United States of America (USA). DCs provide such support; however, if an unplanned disruption (like a natural hazard or power outage) occurs, the functionality of DCs are in jeopardy. Unplanned downtime in DCs cause severe economic and social repercussions. With the Local Indicator of Spatial Association (LISA) test, the research found that there are a large percentage of DCs that are in non-vulnerable areas of disruption; however, there is still a notable percentage in disruption prone areas. For example, earthquakes, hurricanes, and tornadoes have the most DCs in vulnerable areas. After identifying these vulnerabilities, the research identified areas within the USA that have minimal vulnerabilities to both the aforementioned natural hazards and power outages with the BI-LISA test. After doing a composite vulnerability score on the Cold-Spots from the BILISA analysis, the research found three counties with the low vulnerability scores. These are Koochiching, Minnesota (0.091), Schoolcraft, Michigan (0.095), and Houghton, Michigan (0.096).

cs.CY↗

arXiv2024

Bright dipolar excitons in twisted black phosphorus homostructures

Bright dipolar excitons, which contain electrical dipoles and have high oscillator strength, are an ideal platform for studying correlated quantum phenomena. They usually rely on carrier tunneling between two quantum wells or two layers to hybridize with nondipolar excitons to gain oscillator strength. In this work, we uncovered a new type of bright infrared dipolar exciton by stacking 90{\deg}-twisted black phosphorus (BP) structures. These excitons, inherent to the reconstructed band structure, exhibit high oscillator strength. Most importantly, they inherit the linear polarization from BP, which allows light polarization to be used to select the dipole direction. Moreover, the dipole moment and resonance energy can be widely tuned by the thickness of the BP. Our results demonstrate a useful platform for exploring tunable correlated dipolar excitons.

cond-mat.mes-hall↗

arXiv2024

TabDPT: Scaling Tabular Foundation Models on Real Data

Tabular data is one of the most ubiquitous sources of information worldwide, spanning a wide variety of domains. This inherent heterogeneity has slowed the development of Tabular Foundation Models (TFMs) capable of fast generalization to unseen datasets. In-Context Learning (ICL) has recently emerged as a promising solution for TFMs, enabling dynamic adaptation to new tasks without additional tuning. While many studies have attempted to re-purpose large language models for tabular ICL, they have had limited success, so recent works have focused on developing tabular-specific foundation models. In this work, we propose an approach to combine ICL-based retrieval with self supervised learning to train tabular foundation models. We also investigate the utility of real vs. synthetic data for model pre-training, and show that real data can contain useful signal not easily captured in synthetic training. Specifically, we show that incorporating real data during the pre-training phase can lead to significantly faster training and better downstream generalization to unseen data. Our resulting model, TabDPT, achieves strong performance on both regression (CTR23) and classification (CC18) benchmarks. Importantly, we also demonstrate that with our pre-training procedure, scaling both model and data size leads to consistent performance improvements that follow power laws. This echoes scaling laws in LLMs and other foundation models, and suggests that large-scale TFMs can be achievable. We open-source our full pipeline: inference code including trained model weights can be found at github.com/layer6ai-labs/TabDPT-inference, and the training code to reproduce experiments can be found at github.com/layer6ai-labs/TabDPT-training.

cs.LG↗

arXiv2024

Establishing Nationwide Power System Vulnerability Index across US Counties Using Interpretable Machine Learning

Power outages have become increasingly frequent, intense, and prolonged in the US due to climate change, aging electrical grids, and rising energy demand. However, largely due to the absence of granular spatiotemporal outage data, we lack data-driven evidence and analytics-based metrics to quantify power system vulnerability. This limitation has hindered the ability to effectively evaluate and address vulnerability to power outages in US communities. Here, we collected ~179 million power outage records at 15-minute intervals across 3022 US contiguous counties (96.15% of the area) from 2014 to 2023. We developed a power system vulnerability assessment framework based on three dimensions (intensity, frequency, and duration) and applied interpretable machine learning models (XGBoost and SHAP) to compute Power System Vulnerability Index (PSVI) at the county level. Our analysis reveals a consistent increase in power system vulnerability over the past decade. We identified 318 counties across 45 states as hotspots for high power system vulnerability, particularly in the West Coast (California and Washington), the East Coast (Florida and the Northeast area), the Great Lakes megalopolis (Chicago-Detroit metropolitan areas), and the Gulf of Mexico (Texas). Heterogeneity analysis indicates that urban counties, counties with interconnected grids, and states with high solar generation exhibit significantly higher vulnerability. Our results highlight the significance of the proposed PSVI for evaluating the vulnerability of communities to power outages. The findings underscore the widespread and pervasive impact of power outages across the country and offer crucial insights to support infrastructure operators, policymakers, and emergency managers in formulating policies and programs aimed at enhancing the resilience of the US power infrastructure.

cs.CY↗