Zero-shot traffic flow prediction with foundation models: a comparison with deep learning approaches
Yue Li, Qunshan Zhao, Mingshu Wang · 2026
Read Zero-shot traffic flow prediction with foundation models: a comparison with deep learning approaches by Yue Li, Qunshan Zhao, Mingshu Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13328556.txt.
Description
Forecasting traffic flow usually means training a deep network on a city's own history, a cost small operators cannot always pay. Yue Li and two co-authors ask in 2026 whether general foundation models, pretrained on vast time series, can forecast traffic cold, with no local training at all. They line up two such models against trained deep learning rivals.
The contest tests whether broad pretraining can substitute for local data and tuning. Foundation models have lately startled forecasters in other domains, and transport is a natural next arena. The authors frame the study as a practical buying question for traffic managers. Its answer decides whether prediction becomes a service anyone can switch on.
Catalog record
| CREATOR | Yue Li, Qunshan Zhao, Mingshu Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |