SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.08942 · ICLR · 2025

Language in the Flow of Time: Time-Series-Paired Texts Weaved into a Unified Temporal Narrative

Zhining Liu, Jingrui He, Ziwei Wu, Hanghang Tong, Yada Zhu, Zihao Li, Dongqi Fu, Xiao Lin, Jiaru Zou, Lecheng Zheng, Hendrik Hamann

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
idea-isail-lab-uiuc/tats canonical 0 of 2
iDEA-iSAIL-Lab-UIUC/TaTS — 2 of 3
FunctionStatusWhere it lives
HiPPO_LegT Ran iDEA-iSAIL-Lab-UIUC/TaTS/models/FiLM.py
code served (permissive licence) · get_code("80b8de794432949f")
SpectralConv1d Ran iDEA-iSAIL-Lab-UIUC/TaTS/models/FiLM.py
code served (permissive licence) · get_code("de04568bd359bc97")
Model Not yet run iDEA-iSAIL-Lab-UIUC/TaTS/models/FiLM.py
code served (permissive licence) · get_code("31c83e9eea158cf0")
time_features Not yet run idea-isail-lab-uiuc/tats/utils/timefeatures.py
code served (permissive licence) · get_code("9a5fcd4ebfc55d03")
time_features_from_frequency_str Not yet run idea-isail-lab-uiuc/tats/utils/timefeatures.py
code served (permissive licence) · get_code("f8544563682146e5")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

While many advances in time series models focus exclusively on numerical data, research on multimodal time series, particularly those involving contextual textual information, remains in its infancy. With recent progress in large language models and time series learning, we revisit the integration of paired texts with time series through the Platonic Representation Hypothesis (Huh et al., 2024), which posits that representations of different modalities converge to shared spaces. In this context, we identify that time-series-paired texts may naturally exhibit periodic properties that closely mirror those of the original time series. Building on this insight, we propose a novel framework, Texts as Time Series (TaTS), which considers the time-series-paired texts to be auxiliary variables of the time series. TaTS can be plugged into any existing numerical-only time series models and effectively enable them to handle time series data with paired texts. Through extensive experiments on both multimodal time series forecasting and imputation tasks across benchmark datasets with various existing time series models, we demonstrate that TaTS can enhance multimodal predictive performance without modifying model architectures. Our Code is available at https://github. com/iDEA-iSAIL-Lab-UIUC/TaTS.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2502.08942")
get_code_for_paper("2502.08942")
have("2502.08942")

Connect an agent — have() is free.