The field of geo-localization and visual place recognition is rapidly advancing with the integration of multimodal large language models and retrieval-augmented generation. These approaches have demonstrated state-of-the-art performance in estimating geolocations and recognizing places, eliminating the need for expensive fine-tuning or retraining and scaling seamlessly to incorporate new data sources. Furthermore, the introduction of concept-aware alignment modules and test-time scaling frameworks has enhanced interpretability and efficiency in geo-localization tasks. Noteworthy papers in this area include: Street-Level Geolocalization Using Multimodal Large Language Models and Retrieval-Augmented Generation, which achieves higher accuracy compared to traditional methods, and Towards Interpretable Geo-localization, which proposes a novel framework integrating global geo-localization with concept bottlenecks. Additionally, Scale, Don't Fine-tune: Guiding Multimodal LLMs for Efficient Visual Place Recognition at Test-Time presents a zero-shot framework employing Test-Time Scaling that leverages MLLMs' vision-language alignment capabilities. GeoArena: An Open Platform for Benchmarking Large Vision-language Models on WorldWide Image Geolocalization provides a first open platform for evaluating LVLMs on worldwide image geolocalization tasks, offering true in-the-wild and human-centered benchmarking.