GenAI4MoDA@MDM 2025

Generative AI, a subset within the realm of artificial intelligence, has undergone significant progress in recent years. This progress empowers machines to craft, imitate, and produce content that closely mirrors human creations.

The application of generative AI on mobility data can benefit studies on urban and mobility data in several aspects: for instance, researchers can leverage new datasets, perform what-if analysis in cases of changes in, e.g., the road network or on the public transport schedule, and possibly evaluate differences between cities.

Furthermore, generating synthetic mobility datasets can offer numerous benefits regarding privacy, confidentiality, and proprietary concerns.

While the application of these technologies in handling urban and mobility data is still in its early stages, its potential impact on research endeavors and decision support systems for policymakers within the urban context is noteworthy.

The second workshop on Generative AI for Mobility data (GenAI4MoDa) aims to build on the results of the first edition and continue uniting researchers and practitioners to exchange insights into the current state of research on Generative AI for urban data.  

The field of generative AI is rapidly advancing, with significant improvements and paradigm shifts occurring every few months. Since the first edition of the GenAI4MoDa workshop, increasingly complex models have been introduced, aiming to unlock more powerful and practical emergent properties. Additionally, numerous open-source large language models (LLMs) have been proposed, competing with their proprietary counterparts. A new paradigm has also emerged, emphasizing extended inference time (also known as test-time compute) to enable complex reasoning during prompt processing, proving to be an exceptionally powerful approach. 

It is easy to imagine that these changes had a profound and ongoing impact on researchers working with generative AI for mobility data. So, the overarching goal of the 2nd GenAI4MoDa workshop is to continue fostering collaboration and establishing a research network that accelerates the development of novel ideas and practical solutions in this field.