T-LLM: Teaching Large Language Models to Forecast Time Series via Temporal Distillation DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/2018/06 CCS: Computing methodologies Temporal reasoning
Suhan Guo, Bingxu Wang, Shaodan Zhang, Furao Shen · 2026
Read T-LLM: Teaching Large Language Models to Forecast Time Series via Temporal Distillation DOI: XXXXXXX.XXXXXXX Conference: Make sure to enter the correct conference title from your rights confirmation email; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/2018/06 CCS: Computing methodologies Temporal reasoning by Suhan Guo, Bingxu Wang, Shaodan Zhang, Furao Shen (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/arxiv-2602-01937.txt.
Description
Suhan Guo and colleagues teach language models to predict time series with T-LLM, a temporal distillation framework. Time series follows evolving processes, growing only with time, curbing pretraining alone. Existing methods use representation alignment or inference time modules rather than teaching ahead directly. T-LLM transfers predictive behavior from a light teacher during training, combining trend and frequency analysis. The teacher is removed at inference, leaving the model sole forecaster. Benchmark and infectious disease tests show T-LLM beating current methods in full, few and zero shot settings.
Catalog record
| CREATOR | Suhan Guo, Bingxu Wang, Shaodan Zhang, Furao Shen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64591 |
| LICENSE | http://creativecommons.org/licenses/by-sa/4.0/ |
| RIGHTS | CC BY SA |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01937