TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models
The paper titled 'TSFMAudit' addresses the issue of data contamination in time series foundation models (TSFMs). It introduces a method for auditing pretraining contamination, which can lead to overly optimistic performance estimates. The authors evaluate their approach on multiple datasets and compare it against existing baselines.
- ▪Time series foundation models are pretrained on large datasets, raising concerns about contamination in evaluation datasets.
- ▪The proposed method, TSFMAudit, is based on probe adaptation dynamics to identify contamination.
- ▪The study evaluates TSFMAudit on 6 TSFMs and 187 datasets, using documented training source evidence for supervision.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Machine Learning arXiv:2605.26161 (cs) [Submitted on 24 May 2026] Title:TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models Authors:Hongkai Li, Shifeng Xie, Lefei Shen, Zhuo Li, Mouxiang Chen, Xiaobin Zhang, Han Fu, Jianling Sun, Xiaoxue Ren, Chenghao Liu View a PDF of the paper titled TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models, by Hongkai Li and 9 other authors View PDF HTML (experimental) Abstract:Time series foundation models (TSFMs) are increasingly pretrained on large corpora, raising concerns that evaluation datasets may have been exposed during pretraining and thus yield overly optimistic performance estimates.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
