arXiv subjects

Heng Cai

Publications and source records attributed to Heng Cai.

13 recordsLinked to original sources

arXiv2026

BEACON: Behavioral and Semantic Enrichment of AlphaEarth Embeddings through Tri-Modal Contrastive Learning

Geospatial foundation models such as the AlphaEarth Foundation produce compact and globally consistent representations of the Earth's surface that transfer effectively to a wide range of downstream tasks. However, because these models are trained primarily on Earth-observation imagery, their embeddings mainly capture physical and spectral characteristics while encoding human activity and urban function only weakly. To address this limitation, we propose BEACON, a tri-modal contrastive learning framework that aligns three complementary views of urban space: physical representations from AE embeddings, semantic representations from point-of-interest (POI) text, and human behavioral representations from hourly POI visitation, while keeping the deployed representation image-only. Using the Houston Metropolitan Area as a case study area, we evaluated the performance of the BEACON framework on nine downstream tasks, including seven regression and two classification tasks against six baselines (raw coordinates, Space2Vec, SatCLIP, TESSERA, Clay and AlphaEarth), using frozen linear and MLP probes over five seeds. Under a linear probe, BEACON improves relative R^2 over AlphaEarth by up to 43% for obesity prevalence, 34% for poor mental health, and 22% for median household income, while remaining competitive in the prediction of physical and environmental variables. These findings highlight the value of augmenting geospatial foundation models with semantic and behavioral signals, extending their applicability from physical Earth observation to human-centered urban analytics.

cs.LG↗

arXiv2026

DamageArbiter: A Multimodal Arbitration Framework for Disaster Damage Assessment from Street-View Imagery

Analyzing street-view imagery with computer vision models offers a promising approach for rapid, hyperlocal disaster damage assessment, but existing approaches typically rely on black-box pre-trained vision models, which lack interpretability and reliability. This study proposes DamageArbiter, a multimodal disagreement-driven arbitration framework designed to improve the accuracy and reliability of street-view-based damage assessment. DamageArbiter leverages the complementary strengths of unimodal and multimodal models and employs a lightweight logistic regression meta-classifier to arbitrate cases in which model predictions disagree. Using 2,556 post-disaster street-view images, paired with manually generated or large language model (LLM)-generated text descriptions, we systematically compared DamageArbiter with fine-tuned unimodal (image-only and text-only) models and CLIP-based multimodal models in terms of classification performance and overconfidence errors. Results show that DamageArbiter improved accuracy to 75.85% and the Matthews correlation coefficient (MCC) to 0.6188, compared with the best-performing text-only baseline (63.07% accuracy, 0.4126 MCC), image-only baseline (74.33% accuracy, 0.5947 MCC), and CLIP baseline (74.22% accuracy, 0.5915 MCC). The overconfidence analysis further reveals that DamageArbiter substantially reduced the overconfidence error from 70.58% for the best-performing baseline, the image-only ViT model, to 16.45%. Overall, this study demonstrates that accuracy alone is insufficient for evaluating disaster damage classification models and highlights the importance of measuring overconfidence errors as part of model reliability assessment. DamageArbiter thus offers a more reliable framework for rapid, hyperlocal disaster damage assessment from street-view imagery.

cs.CV↗

arXiv2026

Predicting Healthcare System Visitation Flow by Integrating Hospital Attributes and Population Socioeconomics with Human Mobility Data

Healthcare visitation patterns are influenced by a complex interplay of hospital attributes, population socioeconomics, and spatial factors. However, existing research often adopts a fragmented approach, examining these determinants in isolation. This study addresses this gap by integrating hospital capacities, occupancy rates, reputation, and popularity with population SES and spatial mobility patterns to predict visitation flows and analyze influencing factors. Utilizing four years of SafeGraph mobility data and user experience data from Google Maps Reviews, five flow prediction models, Naive Regression, Gradient Boosting, Multilayer Perceptrons (MLPs), Deep Gravity, and Heterogeneous Graph Neural Networks (HGNN),were trained and applied to simulate visitation flows in Houston, Texas, U.S. The Shapley additive explanation (SHAP) analysis and the Partial Dependence Plot (PDP) method were employed to examine the combined impacts of different factors on visitation patterns. The findings reveal that Deep Gravity outperformed other models. Hospital capacities, ICU occupancy rates, ratings, and popularity significantly influence visitation patterns, with their effects varying across different travel distances. Short-distance visits are primarily driven by convenience, whereas long-distance visits are influenced by hospital ratings. White-majority areas exhibited lower sensitivity to hospital ratings for short-distance visits, while Asian populations and those with higher education levels prioritized hospital rating in their visitation decisions. SES further influence these patterns, as areas with higher proportions of Hispanic, Black, under-18, and over-65 populations tend to have more frequent hospital visits, potentially reflecting greater healthcare needs or limited access to alternative medical services.

cs.LG↗

arXiv2025

Toward building next-generation Geocoding systems: a systematic review

Geocoding systems are widely used in both scientific research for spatial analysis and everyday life through location-based services. The quality of geocoded data significantly impacts subsequent processes and applications, underscoring the need for next-generation systems. In response to this demand, this review first characterizes the technical requirements for next-generation geocoding inputs and outputs. We then decompose the geocoding workflow into modular functional units and survey existing implementations. For each component, we identify methodological limitations, articulate domain-specific research questions and hypotheses, and outline evaluation strategies needed. Finally, we identify opportunities to improve next-generation geocoding systems in light of recent technological advances. We envision that this review provides a technical foundation and research agenda for advancing the design, assessment, and deployment of next-generation geocoding systems.

cs.SE↗

arXiv2024

PRIME: A CyberGIS Platform for Resilience Inference Measurement and Enhancement

In an era of increased climatic disasters, there is an urgent need to develop reliable frameworks and tools for evaluating and improving community resilience to climatic hazards at multiple geographical and temporal scales. Defining and quantifying resilience in the social domain is relatively subjective due to the intricate interplay of socioeconomic factors with disaster resilience. Meanwhile, there is a lack of computationally rigorous, user-friendly tools that can support customized resilience assessment considering local conditions. This study aims to address these gaps through the power of CyberGIS with three objectives: 1) To develop an empirically validated disaster resilience model - Customized Resilience Inference Measurement designed for multi-scale community resilience assessment and influential socioeconomic factors identification, 2) To implement a Platform for Resilience Inference Measurement and Enhancement module in the CyberGISX platform backed by high-performance computing, 3) To demonstrate the utility of PRIME through a representative study. CRIM generates vulnerability, adaptability, and overall resilience scores derived from empirical hazard parameters. Computationally intensive Machine Learning methods are employed to explain the intricate relationships between these scores and socioeconomic driving factors. PRIME provides a web-based notebook interface guiding users to select study areas, configure parameters, calculate and geo-visualize resilience scores, and interpret socioeconomic factors shaping resilience capacities. A representative study showcases the efficiency of the platform while explaining how the visual results obtained may be interpreted. The essence of this work lies in its comprehensive architecture that encapsulates the requisite data, analytical and geo-visualization functions, and ML models for resilience assessment.

cs.LG↗

arXiv2024

Understanding Human-COVID-19 Dynamics using Geospatial Big Data: A Systematic Literature Review

The COVID-19 pandemic has changed human life. To mitigate the pandemic's impacts, different regions implemented various policies to contain COVID-19 and residents showed diverse responses. These human responses in turn shaped the uneven spatial-temporal spread of COVID-19. Consequently, the human-pandemic interaction is complex, dynamic, and interconnected. Delineating the reciprocal effects between human society and the pandemic is imperative for mitigating risks from future epidemics. Geospatial big data acquired through mobile applications and sensor networks have facilitated near-real-time tracking and assessment of human responses to the pandemic, enabling a surge in researching human-pandemic interactions. However, these investigations involve inconsistent data sources, human activity indicators, relationship detection models, and analysis methods, leading to a fragmented understanding of human-pandemic dynamics. To assess the current state of human-pandemic interactions research, we conducted a synthesis study based on 67 selected publications between March 2020 and January 2023. We extracted key information from each article across six categories, e.g., research area and time, data, methodological framework, and results and conclusions. Results reveal that regression models were predominant in relationship detection, featured in 67.16% of papers. Only two papers employed spatial-temporal models, notably underrepresented in the existing literature. Studies examining the effects of policies and human mobility on the pandemic's health impacts were the most prevalent, each comprising 12 articles (17.91%). Only 3 papers (4.48%) delved into bidirectional interactions between human responses and the COVID-19 spread. These findings shed light on the need for future research to spatially and temporally model the long-term, bidirectional causal relationships within human-pandemic systems.

physics.soc-ph↗

arXiv2023

CCMB: A Large-scale Chinese Cross-modal Benchmark

Vision-language pre-training (VLP) on large-scale datasets has shown premier performance on various downstream tasks. In contrast to plenty of available benchmarks with English corpus, large-scale pre-training datasets and downstream datasets with Chinese corpus remain largely unexplored. In this work, we build a large-scale high-quality Chinese Cross-Modal Benchmark named CCMB for the research community, which contains the currently largest public pre-training dataset Zero and five human-annotated fine-tuning datasets for downstream tasks. Zero contains 250 million images paired with 750 million text descriptions, plus two of the five fine-tuning datasets are also currently the largest ones for Chinese cross-modal downstream tasks. Along with the CCMB, we also develop a VLP framework named R2D2, applying a pre-Ranking + Ranking strategy to learn powerful vision-language representations and a two-way distillation method (i.e., target-guided Distillation and feature-guided Distillation) to further enhance the learning capability. With the Zero and the R2D2 VLP framework, we achieve state-of-the-art performance on twelve downstream datasets from five broad categories of tasks including image-text retrieval, image-text matching, image caption, text-to-image generation, and zero-shot image classification. The datasets, models, and codes are available at https://github.com/yuxie11/R2D2

cs.CV↗

arXiv2023

Sensing the Pulse of the Pandemic: Geovisualizing the Demographic Disparities of Public Sentiment toward COVID-19 through Social Media

Social media offers a unique lens to observe large-scale, spatial-temporal patterns of users reactions toward critical events. However, social media use varies across demographics, with younger users being more prevalent compared to older populations. This difference introduces biases in data representativeness, and analysis based on social media without proper adjustment will lead to overlooking the voices of digitally marginalized communities and inaccurate estimations. This study explores solutions to pinpoint and alleviate the demographic biases in social media analysis through a case study estimating the public sentiment about COVID-19 using Twitter data. We analyzed the pandemic-related Twitter data in the U.S. during 2020-2021 to (1) elucidate the uneven social media usage among demographic groups and the disparities of their sentiments toward COVID-19, (2) construct an adjusted public sentiment measurement based on social media, the Sentiment Adjusted by Demographics (SAD) index, to evaluate the spatiotemporal varying public sentiment toward COVID-19. The results show higher proportions of female and adolescent Twitter users expressing negative emotions to COVID-19. The SAD index unveils that the public sentiment toward COVID-19 was most negative in January and February 2020 and most positive in April 2020. Vermont and Wyoming were the most positive and negative states toward COVID-19.

cs.CY↗

arXiv2023

3D Medical Image Segmentation with Sparse Annotation via Cross-Teaching between 3D and 2D Networks

Medical image segmentation typically necessitates a large and precisely annotated dataset. However, obtaining pixel-wise annotation is a labor-intensive task that requires significant effort from domain experts, making it challenging to obtain in practical clinical scenarios. In such situations, reducing the amount of annotation required is a more practical approach. One feasible direction is sparse annotation, which involves annotating only a few slices, and has several advantages over traditional weak annotation methods such as bounding boxes and scribbles, as it preserves exact boundaries. However, learning from sparse annotation is challenging due to the scarcity of supervision signals. To address this issue, we propose a framework that can robustly learn from sparse annotation using the cross-teaching of both 3D and 2D networks. Considering the characteristic of these networks, we develop two pseudo label selection strategies, which are hard-soft confidence threshold and consistent label fusion. Our experimental results on the MMWHS dataset demonstrate that our method outperforms the state-of-the-art (SOTA) semi-supervised segmentation methods. Moreover, our approach achieves results that are comparable to the fully-supervised upper bound result.

cs.CV↗

arXiv2023

Orthogonal Annotation Benefits Barely-supervised Medical Image Segmentation

Recent trends in semi-supervised learning have significantly boosted the performance of 3D semi-supervised medical image segmentation. Compared with 2D images, 3D medical volumes involve information from different directions, e.g., transverse, sagittal, and coronal planes, so as to naturally provide complementary views. These complementary views and the intrinsic similarity among adjacent 3D slices inspire us to develop a novel annotation way and its corresponding semi-supervised model for effective segmentation. Specifically, we firstly propose the orthogonal annotation by only labeling two orthogonal slices in a labeled volume, which significantly relieves the burden of annotation. Then, we perform registration to obtain the initial pseudo labels for sparsely labeled volumes. Subsequently, by introducing unlabeled volumes, we propose a dual-network paradigm named Dense-Sparse Co-training (DeSCO) that exploits dense pseudo labels in early stage and sparse labels in later stage and meanwhile forces consistent output of two networks. Experimental results on three benchmark datasets validated our effectiveness in performance and efficiency in annotation. For example, with only 10 annotated slices, our method reaches a Dice up to 86.93% on KiTS19 dataset.

cs.CV↗

arXiv2021

Social Media for Emergency Rescue: An Analysis of Rescue Requests on Twitter during Hurricane Harvey

Social media plays increasingly significant roles in disaster response, but effectively leveraging social media for rescue is challenging. This study analyzed rescue requests on Twitter during the 2017 Hurricane Harvey, in which many residents resorted to social media to call for help. The objectives include (1) understanding the characteristics of rescue-request messages; (2) revealing the spatial-temporal patterns of rescue requests; (3) determining the social-geographical conditions of communities needing rescue; and (4) identifying the challenges of using social media for rescue and propose improvement strategies. About half of rescue requests either did not provide sufficient information or neglected to include rescue-related hashtags or accounts. Of the 824 geocoded unique rescue requests, 41% were from FEMA-defined minimal flood risk zones. Communities sending more rescue requests on Twitter were environmentally and socioeconomically more vulnerable. Finally, we derived a framework summarizing the steps and strategies needed to improve social media use for rescue operations.

cs.SI↗

arXiv2021

Revealing the Global Linguistic and Geographical Disparities of Public Awareness to Covid-19 Outbreak through Social Media

The Covid-19 has presented an unprecedented challenge to public health worldwide. However, residents in different countries showed diverse levels of Covid-19 awareness during the outbreak and suffered from uneven health impacts. This study analyzed the global Twitter data from January 1st to June 30th, 2020, seeking to answer two research questions. What are the linguistic and geographical disparities of public awareness in the Covid-19 outbreak period reflected on social media? Can the changing pandemic awareness predict the Covid-19 outbreak? We established a Twitter data mining framework calculating the Ratio index to quantify and track the awareness. The lag correlations between awareness and health impacts were examined at global and country levels. Results show that users presenting the highest Covid-19 awareness were mainly those tweeting in the official languages of India and Bangladesh. Asian countries showed more significant disparities in awareness than European countries, and awareness in the eastern part of Europe was higher than in central Europe. Finally, the Ratio index could accurately predict global mortality rate, global case fatality ratio, and country-level mortality rate, with 21-30, 35-42, and 17 leading days, respectively. This study yields timely insights into social media use in understanding human behaviors for public health research.

physics.soc-ph↗

arXiv2016

Linear absorption coefficient of in-plane graphene on a silicon microring resonator

We demonstrate that linear absorption coefficient (LAC) of a graphene-silicon hybrid waveguide (GSHW) is determined by the optical transmission spectra of a graphene coated symmetrically coupled add-drop silicon microring resonator (SC-ADSMR), of which the value is around 0.23 dB/um. In contrast to the traditional cut-back method, the measured results are not dependent on the coupling efficiency of the fiber tip and the waveguide. Moreover, precision evaluation of graphene coated silicon microring resonator (SMR) is crucial for the optoelectronic devices targeting for compact footprint and low power consumption.

physics.optics↗