AI Paper Atlas

Research map / Time series / Foundation models for time series

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

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

ICLR 2026 · Time series › Foundation models for time series · approach: Time-series language reasoning · 33 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add paired texts as auxiliary variables (TaTS) to iTransformeriTransformer · Time-MMD EnvironmentMSE: 0.441 -> 0.267Improved
add paired texts as auxiliary variables (TaTS) to PatchTSTPatchTST · Time-MMD EconomyMSE: 0.017 -> 0.009Improved
use TaTS instead of MM-TSFLib text fusioniTransformer · Time-MMD EnvironmentMSE: 0.421 -> 0.267Improved
add paired texts via TaTSiTransformer · Time-MMD EnergyMAE: 0.375 -> 0.376No clear effect
add TaTS to imputationPatchTST · Time-MMD Traffic imputationMSE: 0.210 -> 0.131Improved
replace MLP projection with gated residual fusion (ablation)TaTS · Time-MMD ClimateMSE: 0.992 -> 0.998No clear effect
replace MLP projection with cross-attention fusion (ablation)TaTS · Time-MMD ClimateMSE: 0.992 -> 0.989No clear effect
keep texts aligned with timestamps instead of shuffling them (ablation)iTransformer + TaTS · Time-MMD ClimateMSE: 1.242 -> 1.028Improved
keep texts aligned with timestamps instead of shuffling them (ablation)iTransformer + TaTS · Time-MMD TrafficMSE: 0.223 -> 0.187Improved
avoid randomly dropping 25% of texts (ablation)iTransformer + TaTS · Time-MMD TrafficMSE: 0.203 -> 0.187Improved
drop part of extremely noisy texts (20% corrupted instead of fully corrupted) (ablation)iTransformer + TaTS · Time-MMD ClimateMSE: 1.242 -> 1.201Improved
use a larger text encoder (BERT 110M to GPT-2 to LLaMA2 7B) (ablation)TaTS · Time-MMD datasetsMSEImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy