LE4Mob is proposed, an inductive, distance-aware, and geography-derived location embedding framework for mobility modelling that extends contrastive language-location pre-training while introducing a distance-aware regularisation objective that encourages the embedding space to preserve spatial relationships.
Abstract
Location representations provide mobility models with fundamental information about the spatial position, functional characteristics, and relationships of places. However, existing embeddings are often dependent on mobility observations, unable to represent unseen locations, and weakly constrained to retain geographic distance. This limits their reuse across datasets and mobility tasks. To address these limitations, we propose LE4Mob, an inductive, distance-aware, and geography-derived location embedding framework for mobility modelling. LE4Mob extends contrastive language-location pre-training while introducing a distance-aware regularisation objective that encourages the embedding space to preserve spatial relationships. Pre-trained from geographic context, LE4Mob can encode rich spatial-semantic information and generate embeddings for unseen locations inductively. Its independence from downstream mobility task supervision also makes it transferable across different mobility tasks. We evaluate LE4Mob on individual-level next location prediction and population-level commuter flow generation. Experiments across multiple datasets and study areas show that LE4Mob outperforms strong baselines, with particular advantages in inductive settings and when downstream models rely directly on interactions between location embeddings. These findings demonstrate the potential of distance-aware, geography-derived location representations as reusable foundations for human mobility modelling.
Experiments show that MoRAL consistently outperforms statistical and LLM-based baselines under the GEO-BLEU metric, improving the mean score over the strongest baseline by 5.82% and achieving a 2.6 times reduction in per-user inference time relative to recent LLM baselines using the same backbone.
Chenhao Wang, Qingsen Ge, Nan Hu et al.· GeoInformatica· 0 citations
This work proposes a lightweight, model-agnostic approach for injecting geo-spatial awareness into pretrained embeddings without modifying the tokenizer or requiring costly retraining, and generalizes across multiple model architectures and scales.
Accurate next-location prediction remains a challenging problem in spatio-temporal mobility modeling. Many existing approaches rely primarily on raw trajectory sequences and overlook the semantic and contextual factors that govern human movement behavior. To address this limitation, this study proposes a novel Spatio-T...
H. Simanjuntak, Marcel Joshua Parulian Simanjuntak, Ronycholson Sinaga et al.· Journal of Artificial Intell...· 0 citations
Spatially-Intrinsic SID (SI-SID), a unified framework whose three synergistic stages jointly construct, cultivate, and preserve urban spatial knowledge, is proposed, which demonstrates the effectiveness of SPAR.
Fang-Ye Wang, Yun-Jin Gu, Haowen Lin et al.· 0 citations
This work introduces PlacePulse, a structural geospatial foundation model that compresses 572 attributes, expanded to 2,288 inputs through H3 neighborhood aggregates, into 256-dimensional unit-norm embeddings for 1,055,537 H3 resolution-7 cells in the United States, and introduces a reusable two-axis benchmark that sep...
Lucía García-Duarte, M. Álvarez, D. Vicente et al.· 0 citations
The results provide evidence that output embeddings encode signals relevant to time and space, while also offering a practical tool for interpretability and for downstream temporal and geographic information retrieval tasks, such as temporal ordering, geographic ranking, and tagging.
Esteban Feuerstein, Victoria Klimkowski, J. O. D. de Zarate et al.· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026