As large language models (LLMs) continue to advance and gain widespread use, establishing systematic and reliable evaluation methodologies for LLMs and vision-language models (VLMs) has become essential to ensure their real-world effectiveness and reliability. There have been some early explorations about the usability of LLMs for limited urban tasks, but a systematic and scalable evaluation benchmark is still lacking. The challenge in constructing a systematic evaluation benchmark for urban research lies in the diversity of urban data, the complexity of application scenarios and the highly dynamic nature of the urban environment. In this paper, we design CityBench, an interactive simulator based evaluation platform, as the first systematic benchmark for evaluating the capabilities of LLMs for diverse tasks in urban research. First, we build CityData to integrate the diverse urban data and CitySimu to simulate fine-grained urban dynamics. Based on CityData and CitySimu, we design 8 representative urban tasks in 2 categories of perception-understanding and decision-making as the CityBench. With extensive results from 30 well-known LLMs and VLMs in 13 cities around the world, we find that advanced LLMs and VLMs can achieve competitive performance in diverse urban tasks requiring commonsense and semantic understanding abilities, e.g., understanding the human dynamics and semantic inference of urban images. Meanwhile, they fail to solve the challenging urban tasks requiring professional knowledge and high-level numerical abilities, e.g., geospatial prediction and traffic control task. These findings provide critical insights for the effective utilization and further development of LLMs to advance urban-related tasks and research in the future.

Large language models(LLMs), with their powerful language generation and reasoning capabilities, have already achieved notable success in many domains, e.g., math and code generation. However, they often fall short when tackling real-life geospatial tasks within urban environments. This limitation stems from a lack of physical world knowledge and relevant data during training. To address this gap, we propose CityGPT, a systematic framework designed to enhance LLMs' understanding of urban space and improve their ability to solve the related urban tasks by integrating a city-scale 'world model' into the model. Firstly, we construct a diverse instruction tuning dataset, CityInstruction, for injecting urban knowledge into LLMs and effectively boosting their spatial reasoning capabilities. Using a combination of CityInstruction and open source general instruction data, we introduce a novel and easy-to-use self-weighted fine-tuning method (SWFT) to train various LLMs (including ChatGLM3-6B, Llama3-8B, and Qwen2.5-7B) to enhance their urban spatial capabilities without compromising, or even improving, their general abilities. Finally, to validate the effectiveness of our proposed framework, we develop a comprehensive text-based spatial benchmark CityEval for evaluating the performance of LLMs across a wide range of urban scenarios and geospatial tasks. Extensive evaluation results demonstrate that smaller LLMs trained with CityInstruction by SWFT method can achieve performance that is competitive with, and in some cases superior to, proprietary LLMs when assessed using CityEval. Our work highlights the potential for integrating spatial knowledge into LLMs, thereby expanding their spatial cognition abilities and applicability to the real-world physical environments. The dataset, benchmark, and source code are open-sourced and can be accessed through https://github.com/tsinghua-fib-lab/CityGPT.

Marker-based optical motion capture (MoCap) is a vital tool in applications such as virtual production, and movement sciences. However, reconstructing scattered MoCap data into real motion sequences is challenging, and data processing is time-consuming and labor-intensive. Here we propose a novel framework for MoCap auto-labeling and matching. In this framework, we designed novel clusters of reflective markers called auto-labeling encoded marker clusters (AEMCs), including clusters with an explicit header (AEMCs-E) and an implicit header (AEMCs-I). Combining cluster design and coding theory gives each cluster a unique codeword for MoCap auto-labeling and matching. Moreover, we provide a method of mapping and decoding for cluster labeling. The labeling results are only determined by the intrinsic characteristics of the clusters instead of the skeleton structure or posture of the subjects. Compared with commercial software and data-driven methods, our method has better labeling accuracy in heterogeneous targets and unknown marker layouts, which demonstrates the promising application of motion capture in humans, rigid or flexible robots.