Can We Predict Your Next Move Without Breaking Your Privacy?
Not on the map. Why?
Abstract (arXiv)
We propose FLLL3M--Federated Learning with Large Language Models for Mobility Modeling--a privacy-preserving framework for Next-Location Prediction (NxLP). By retaining user data locally and leveraging LLMs through an efficient outer product mechanism, FLLL3M ensures high accuracy with low resource demands. It achieves SOT results on Gowalla (Acc@1: 12.55, MRR: 0.1422), WeePlace (10.71, 0.1285), Brightkite (10.42, 0.1169), and FourSquare (8.71, 0.1023), while reducing parameters by up to 45.6% and memory usage by 52.7%.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 8 of its 17 references that are in the map. First posted on arXiv 8 Jul 2025.
Builds on
8 of 17 references.
References · 8
- 2017Attention Is All You Needcited by 60,361
- 2023Spatio-temporal Diffusion Point Processescited by 52
- 2024Efficient World Models with Context-Aware Tokenizationcited by 33
- 2022Region Embedding with Intra and Inter-View Contrastive Learningcited by 24
- 2021Virtual Augmentation Supported Contrastive Learning of Sentence Representationscited by 21
- 2022Learning Temporal Point Processes for Efficient Retrieval of Continuous Time Event Sequencescited by 14
- 2021LightMove: A Lightweight Next-POI Recommendation for Taxicab Rooftop Advertisingcited by 9
- 2019Category-Aware Location Embedding for Point-of-Interest Recommendationcited by 8
Built on by
Not cited yet.