Foundation models (FMs) promise to generalize medical imaging, but their effectiveness varies. It remains unclear how pre-training domain (medical vs. general), paradigm (e.g., text-guided), and architecture influence embedding quality, hindering the selection of optimal encoders for specific radiology tasks. To address this, we evaluate vision encoders from eight medical and general-domain FMs for chest X-ray analysis. We benchmark classification (pneumothorax, cardiomegaly) and segmentation (pneumothorax, cardiac boundary) using linear probing and fine-tuning. Our results show that domain-specific pre-training provides a significant advantage; medical FMs consistently outperformed general-domain models in linear probing, establishing superior initial feature quality. However, feature utility is highly task-dependent. Pre-trained embeddings were strong for global classification and segmenting salient anatomy (e.g., heart). In contrast, for segmenting complex, subtle pathologies (e.g., pneumothorax), all FMs performed poorly without significant fine-tuning, revealing a critical gap in localizing subtle disease. Subgroup analysis showed FMs use confounding shortcuts (e.g., chest tubes for pneumothorax) for classification, a strategy that fails for precise segmentation. We also found that expensive text-image alignment is not a prerequisite; image-only (RAD-DINO) and label-supervised (Ark+) FMs were among top performers. Notably, a supervised, end-to-end baseline remained highly competitive, matching or exceeding the best FMs on segmentation tasks. These findings show that while medical pre-training is beneficial, architectural choices (e.g., multi-scale) are critical, and pre-trained features are not universally effective, especially for complex localization tasks where supervised models remain a strong alternative.

To date, there exist almost no culturally-specific evaluation benchmarks for large language models (LLMs) that cover a large number of languages and cultures. In this paper, we present Global PIQA, a participatory commonsense reasoning benchmark for over 100 languages, constructed by hand by 335 researchers from 65 countries around the world. The 116 language varieties in Global PIQA cover five continents, 14 language families, and 23 writing systems. In the non-parallel split of Global PIQA, over 50% of examples reference local foods, customs, traditions, or other culturally-specific elements. We find that state-of-the-art LLMs perform well on Global PIQA in aggregate, but they exhibit weaker performance in lower-resource languages (up to a 37% accuracy gap, despite random chance at 50%). Open models generally perform worse than proprietary models. Global PIQA highlights that in many languages and cultures, everyday knowledge remains an area for improvement, alongside more widely-discussed capabilities such as complex reasoning and expert knowledge. Beyond its uses for LLM evaluation, we hope that Global PIQA provides a glimpse into the wide diversity of cultures in which human language is embedded.

Foundation models, trained on vast amounts of data using self-supervised techniques, have emerged as a promising frontier for advancing artificial intelligence (AI) applications in medicine. This study evaluates three different vision-language foundation models (RAD-DINO, CheXagent, and BiomedCLIP) on their ability to capture fine-grained imaging features for radiology tasks. The models were assessed across classification, segmentation, and regression tasks for pneumothorax and cardiomegaly on chest radiographs. Self-supervised RAD-DINO consistently excelled in segmentation tasks, while text-supervised CheXagent demonstrated superior classification performance. BiomedCLIP showed inconsistent performance across tasks. A custom segmentation model that integrates global and local features substantially improved performance for all foundation models, particularly for challenging pneumothorax segmentation. The findings highlight that pre-training methodology significantly influences model performance on specific downstream tasks. For fine-grained segmentation tasks, models trained without text supervision performed better, while text-supervised models offered advantages in classification and interpretability. These insights provide guidance for selecting foundation models based on specific clinical applications in radiology.

Visual-Language Models (VLMs) have shown remarkable performance across various tasks, particularly in recognizing geographic information from images. However, significant challenges remain, including biases and privacy concerns. To systematically address these issues in the context of geographic information recognition, we introduce a benchmark dataset consisting of 1,200 images paired with detailed geographic metadata. Evaluating four VLMs, we find that while these models demonstrate the ability to recognize geographic information from images, achieving up to $53.8\%$ accuracy in city prediction, they exhibit significant regional biases. Specifically, performance is substantially higher for economically developed and densely populated regions compared to less developed ($-12.5\%$) and sparsely populated ($-17.0\%$) areas. Moreover, the models exhibit regional biases, frequently overpredicting certain locations; for instance, they consistently predict Sydney for images taken in Australia. The strong performance of VLMs also raises privacy concerns, particularly for users who share images online without the intent of being identified. Our code and dataset are publicly available at https://github.com/uscnlp-lime/FairLocator.

In the small village of Ogba-Ojibo in central Nigeria, sitting at the confluence of two of the nation's largest rivers – the Niger and Benue – 27-year-old Ako Prince Omali is counting the steps carved out of the dirt, which lead down the loam-coloured banks of the river Niger. This river bank, dotted with tufts of spiky grass, is where villagers come to fish or wash produce and laundry. Just last week, three of the steps were submerged during one night of rain, which raised the water level by about five metres. Normally, you can count seven steps down into the river. Now, only four remain above the surface of the water, the sticks bracing the muddy steps having washed away in the deluge.

Spannotation is an open source user-friendly tool developed for image annotation for semantic segmentation specifically in autonomous navigation tasks. This study provides an evaluation of Spannotation, demonstrating its effectiveness in generating accurate segmentation masks for various environments like agricultural crop rows, off-road terrains and urban roads. Unlike other popular annotation tools that requires about 40 seconds to annotate an image for semantic segmentation in a typical navigation task, Spannotation achieves similar result in about 6.03 seconds. The tools utility was validated through the utilization of its generated masks to train a U-Net model which achieved a validation accuracy of 98.27% and mean Intersection Over Union (mIOU) of 96.66%. The accessibility, simple annotation process and no-cost features have all contributed to the adoption of Spannotation evident from its download count of 2098 (as of February 25, 2024) since its launch. Future enhancements of Spannotation aim to broaden its application to complex navigation scenarios and incorporate additional automation functionalities. Given its increasing popularity and promising potential, Spannotation stands as a valuable resource in autonomous navigation and semantic segmentation. For detailed information and access to Spannotation, readers are encouraged to visit the project's GitHub repository at https://github.com/sof-danny/spannotation

The emergence of tethered drones represents a major advancement in unmanned aerial vehicles (UAVs) offering solutions to key limitations faced by traditional drones. This article explores the potential of tethered drones with a particular focus on their ability to tackle issues related to battery life constraints and data latency commonly experienced by battery operated drones. Through their connection to a ground station via a tether, autonomous tethered drones provide continuous power supply and a secure direct data transmission link facilitating prolonged operational durations and real time data transfer. These attributes significantly enhance the effectiveness and dependability of drone missions in scenarios requiring extended surveillance, continuous monitoring and immediate data processing needs. Examining the advancements, operational benefits and potential future progressions associated with tethered drones, this article shows their increasing significance across various sectors and their pivotal role in pushing the boundaries of current UAV capabilities. The emergence of tethered drone technology not only addresses existing obstacles but also paves the way for new innovations within the UAV industry.

Language identification (LID) is a crucial precursor for NLP, especially for mining web data. Problematically, most of the world's 7000+ languages today are not covered by LID technologies. We address this pressing issue for Africa by introducing AfroLID, a neural LID toolkit for $517$ African languages and varieties. AfroLID exploits a multi-domain web dataset manually curated from across 14 language families utilizing five orthographic systems. When evaluated on our blind Test set, AfroLID achieves 95.89 F_1-score. We also compare AfroLID to five existing LID tools that each cover a small number of African languages, finding it to outperform them on most languages. We further show the utility of AfroLID in the wild by testing it on the acutely under-served Twitter domain. Finally, we offer a number of controlled case studies and perform a linguistically-motivated error analysis that allow us to both showcase AfroLID's powerful capabilities and limitations.

In order to ease the process of library management many technologies have been adopted but most of them focus on inventory management. There has hardly been any progress of automation in the field of issuing and returning books to the library on time. In colleges and schools, hostellers often forget to timely return the issued books back to the library. To solve the above issue and to ensure timely submission of the issued books, this work develops a Book-Bot which solves these complexities. The bot can commute from point A to point B, scan and verify QR Codes and Barcodes. The bot will have a certain payload capacity for carrying books. The QR code and Barcode scanning will be enabled by a Pi Camera, OpenCV and Raspberry Pi, thus making the exchange of books safe and secure. The odometry maneuvers of the bot will be controlled manually via a Blynk App. This paper focuses on how human intervention can be reduced and automates the issue part of library management system with the help of a bot.

Collaborative AI systems (CAISs) aim at working together with humans in a shared space to achieve a common goal. This critical setting yields hazardous circumstances that could harm human beings. Thus, building such systems with strong assurances of compliance with requirements, domain-specific standards and regulations is of greatest importance. Only few scale impact has been reported so far for such systems since much work remains to manage possible risks. We identify emerging problems in this context and then we report our vision, as well as the progress of our multidisciplinary research team composed of software/systems, and mechatronics engineers to develop a risk-driven assurance process for CAISs.