Time Series Foundation Models (TSFMs) have recently emerged as a highly promising paradigm for cross-domain zero-shot forecasting. However, existing evaluation protocols predominantly rely on static benchmarks with fixed historical test windows.
LiveHouse-TS introduces an open‑world living benchmark for Time Series Foundation Models, evaluating them prequentially on real future data rather than static test windows. The benchmark captures continuous performance across seasonal changes, distribution shifts, and unexpected events, providing a more realistic assessment of model robustness. Experiments across 11 domains and 17 datasets show that model rankings can dramatically change under this live protocol.
By Haomin Wen, Ziyu Zhou, Qingxiang Liu, Siru Zhong, Yuxuan Liang
arXiv:2511. 03877v2 Announce Type: replace Abstract: Social and collaborative platforms emit multivariate time-series traces in which early interactions -- such as views, likes, or downloads -- are followed, sometimes months or years later, by higher impact like citations, sales, or reviews.
By Kimia Kazemian (Department of Computer Science, Cornell University), Zhenzhen Liu (Department of Computer Science, Cornell University), Yangfanyu Yang (Department of Information Science, Cornell University), Katie Luo (Department of Computer Science, Stanford University), Shuhan Gu (Department of Computer Science, Cornell University), Audrey Du (Department of Computer Science, Cornell University), Xinyu Yang (Department of Information Science, Cornell University), Jack Jansons (Department of Computer Science, Cornell University), Kilian Q. Weinberger (Department of Computer Science, Cornell University), John Thickstun (Department of Computer Science, Cornell University), Yian Yin (Department of Information Science, Cornell University), Sarah Dean (Department of Computer Science, Cornell University)
arXiv:2609.13789v1 Announce Type: new
Abstract: In multi-channel paid user acquisition, early and accurate prediction of user retention at the channel level is crucial for optimizing budget allocatio...
By Zibo Zhao, Zhengxiong Guan, Chaoli Zhang, Linyuan Geng, Xuanbing Zhu, Zhonglong Zheng, Fan Wu
arXiv:2602. 06136v2 Announce Type: replace Abstract: Test-time adaptation (TTA) offers a compelling remedy for machine learning (ML) models that degrade under domain shifts, improving generalisation on-the-fly with only unlabelled samples.
By Sudarshan Sreeram, Young D. Kwon, Cecilia Mascolo
arXiv:2607. 11889v1 Announce Type: cross Abstract: Large language models trained on unrestricted internet corpora inevitably embed information from the future, introducing lookahead bias that compromises the validity of backtests and causal inference in finance and the social sciences.
By Bryan Kelly, Semyon Malamud, Johannes Schwab, Teng Andrea Xu