arXiv cs.CY· Xin Wang, Paraic Carroll, Kerry Nice, Sachith Seneviratne, Li Zhang·· 9 天前AI 评分27
哪些个人信息能提升 LLM 的下一位置预测？
What Personal Information Improves LLM-Based Next-Location Prediction?
AI 导读
一项基于深圳 5000 名居民 sociodemographic 记录与移动轨迹的研究发现，在 GPT-5、GPT-5.5 和 Claude Opus 4.6 的 1000 组配对预测中，行为历史将 top-1 准确率从 5.6% 提升至 18.5%（历史从零增至六天），而年龄、性别、职业、收入等 sociodemographic 属性未带来可检测的整体增益。
正文 · 原文
Abstract:Large language models (LLMs) are increasingly used for individual next-location prediction, with personal information easily added to prompts alongside mobility history. Yet the incremental predictive value of such information remains unclear. Using linked sociodemographic records and mobility traces from 5,000 Shenzhen residents, this study separates model responsiveness from predictive value. GPT-5 is the primary model, with GPT-5.5 and Claude Opus 4.6 used for replication. In 1,000 paired prediction instances, models rank 100 candidate destinations with and without age, gender, occupation and income while all other inputs are held fixed. Behavioural history raises top-1 accuracy from 5.6% to 18.5% as prior history increases from zero to six days. By contrast, sociodemographic attributes produce no detectable overall gain, although replacing correct attributes with those of another person reduces accuracy by 5.4 percentage points. Candidate construction also matters, removing distance raises accuracy by 7.7 points under proximity sampling but lowers it by 22.3 points under popularity sampling, with the reversal reproduced across all three LLMs. These findings identify behavioural history as the clearest source of incremental value and show that personal information should be evaluated under matched, explicitly specified conditions before its privacy and governance costs are justified.
| Comments: | 20 pages, 5 figures |
| Subjects: | Computers and Society (cs.CY) |
| Cite as: | arXiv:2609.09609 [cs.CY] |
| (or arXiv:2609.09609v2 [cs.CY] for this version) | |
| https://doi.org/10.48550/arXiv.2609.09609 arXiv-issued DOI via DataCite |
Submission history
From: Xin Wang [view email]
[v1]
Wed, 9 Sep 2026 02:06:10 UTC (2,106 KB)
[v2]
Thu, 8 Oct 2026 01:55:23 UTC (2,434 KB)
来源：arXiv cs.CY · arxiv.org