Saved in:
Bibliographic Details
Main Authors: Soni, Arpita, Tripathi, Sahil, Kashyap, Gautam Siddharth, Kulahara, Manaswi, Azeez, Mohammad Anas, Siddiqui, Zohaib Hasan, Joshi, Nipun, Gao, Jiechao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.08843
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908447235112960
author Soni, Arpita
Tripathi, Sahil
Kashyap, Gautam Siddharth
Kulahara, Manaswi
Azeez, Mohammad Anas
Siddiqui, Zohaib Hasan
Joshi, Nipun
Gao, Jiechao
author_facet Soni, Arpita
Tripathi, Sahil
Kashyap, Gautam Siddharth
Kulahara, Manaswi
Azeez, Mohammad Anas
Siddiqui, Zohaib Hasan
Joshi, Nipun
Gao, Jiechao
contents 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%.
format Preprint
id arxiv_https___arxiv_org_abs_2507_08843
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Can We Predict Your Next Move Without Breaking Your Privacy?
Soni, Arpita
Tripathi, Sahil
Kashyap, Gautam Siddharth
Kulahara, Manaswi
Azeez, Mohammad Anas
Siddiqui, Zohaib Hasan
Joshi, Nipun
Gao, Jiechao
Machine Learning
Artificial Intelligence
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%.
title Can We Predict Your Next Move Without Breaking Your Privacy?
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2507.08843