Large-scale biodiversity monitoring platforms increasingly rely on multimodal wildlife observations. While recent foundation models enable rich semantic representations across vision, audio, and language, retrieving relevant observations from massive archives remains challenging due to the computational cost of high-dimensional similarity search. In this work, we introduce compact hypercube embeddings for fast text-based wildlife observation retrieval, a framework that enables efficient text-based search over large-scale wildlife image and audio databases using compact binary representations. Building on the cross-view code alignment hashing framework, we extend lightweight hashing beyond a single-modality setup to align natural language descriptions with visual or acoustic observations in a shared Hamming space. Our approach leverages pretrained wildlife foundation models, including BioCLIP and BioLingual, and adapts them efficiently for hashing using parameter-efficient fine-tuning. We evaluate our method on large-scale benchmarks, including iNaturalist2024 for text-to-image retrieval and iNatSounds2024 for text-to-audio retrieval, as well as multiple soundscape datasets to assess robustness under domain shift. Results show that retrieval using discrete hypercube embeddings achieves competitive, and in several cases superior, performance compared to continuous embeddings, while drastically reducing memory and search cost. Moreover, we observe that the hashing objective consistently improves the underlying encoder representations, leading to stronger retrieval and zero-shot generalization. These results demonstrate that binary, language-based retrieval enables scalable and efficient search over large wildlife archives for biodiversity monitoring systems.
Background: Ayush systems form a crucial component of India's healthcare system. Despite sustained policy attention, the implementation of the National Ayush Mission scheme and progress in the sector, challenges persist in the availability and utilization of Ayush services in several States/UTs. Drawing on the National Sample Survey (79th Round, 2022–23) on Ayush, this study examines state-wise variations in awareness, usage, and out-of-pocket expenditure (OOPE) on Ayush medicines, alongside the availability of infrastructure, human resources, and financial burden. Methods: A descriptive cross-sectional analysis was undertaken using NSSO 79th Round data, supplemented with information on Ayush dispensaries and registered practitioners. Population coverage per dispensary and per practitioner served as proxy indicators for infrastructure and workforce availability. States/UTs were categorized by availability levels and usage relative to the national average (496 users per 1000). Cross-tabulations and comparative categorizations were used to examine inter-state variation. Results: National awareness exceeded 95%, while usage remained around 50%, with substantial variation across States/UTs. 20 States/UTs reported above-national average usage and sixteen below. Infrastructure availability varied widely, with only five States/UTs meeting high dispensary coverage and thirteen exhibiting very low coverage. Practitioner availability was similarly uneven, ranging from zero in several Northeastern States to strong density in Kerala and Maharashtra. Higher availability did not consistently correspond with higher usage. OOPE averaged ₹523 per capita annually, with significant inter-state differences, indicating non-linear associations. Recommendations and conclusion: Despite widespread awareness, Ayush utilization remains uneven due to infrastructure gaps and workforce disparities. Strengthening infrastructure and human resources, improving quality of care, expanding digital systems, enhancing public financing, education and training, promoting evidence-informed communication, and reinforcing research–practice linkages are essential to improve access, utilization and financial protection.
The increase in floods in Argentina in recent years is attributed to various factors, including climate change, poor urban planning and inadequate drainage systems. This study examines flood risk in General Rodriguez, Buenos Aires province, and its spatial distribution using the Social Theory of Risk to estimate scenarios forextreme events, with return periods of up to 500 years. The findings were used to develop a local risk map. The methodology involves analyzing hazard through hydrological and hydraulic modeling and constructing aspatial social vulnerability index based on indicators derived from census variables. The combination of thesefactors enables the estimation of urban flood risk, culminating in a risk map that serves as a valuable tool forplanning and efficient resource management.
This study examines the influence of macroeconomic and firm-specific risks on the leverage of publicly listed Indonesian manufacturing and non-financial service firms. It also breaks down the divergence between sectors of risk responsiveness, which, in the capital structure literature for emerging markets, remains mostly uninvestigated. Using a sample of 99 publicly listed firms on the Indonesia Stock Exchange (IDX) from 2010 to 2019, we apply an Instrumental Variable (IV) and system- GMM estimator to control for endogeneity. The findings suggest that increasing macroeconomic risk and reducing firm-specific risk induces leverage, especially among service firms. These results also have policy implications for guiding firms in aligning their finance strategies to the sectoral risk they face and for assisting in formulating tailored policies that maintain robustness in the form of financial industry stability and corporate growth.
