arXiv Computation and Language

TimeOmni-o1: Bidirectionally Coupling Time Series Foundation Models and LLMs for Multimodal Reasoning and Forecasting

TimeOmni-o1 is a new model that bidirectionally couples a pretrained time‑series foundation model (TSFM) with a large language model (LLM) to simultaneously read, generate, and reason over both time‑series data and accompanying text. The TSFM encodes the series for the LLM, which performs multimodal reasoning, and the LLM’s output is fed back to the TSFM to produce forecasts conditioned on that reasoning. Trained on the newly created MMTR dataset—over 500,000 samples across 12 domains—TimeOmni-o1 surpasses existing models on both understanding and forecasting tasks, and its performance generalizes to three open benchmarks.

arXiv Machine Learning
Oct 2

OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning

arXiv:2609.40265v1 Announce Type: new Abstract: Real-world time-series applications increasingly require models that can handle time series forecasting, context-conditioned prediction, and language-b...

By Tony Chen, Timo Stoffregen, Maxwell Xu, Thomas Kaar, Martin Maritsch, Geremia Pompei, Nicolas Zumarraga, Robert Jakob, Paul Schmiedmayer, Patrick Langer, Juncheng Liu
arXiv AI
Sep 11

KairosAgent: Agentic Time Series Forecasting with Fused Semantic Reasoning

KairosAgent is an agentic framework that combines a large language model (LLM) reasoner with a time series foundation model (TSFM) forecaster to tackle cross‑domain multimodal time series forecasting. It dynamically invokes analytical tools to improve the LLM’s numerical comprehension and semantic reasoning, then fuses the reasoning outcomes into the TSFM pipeline for more accurate predictions. The approach is further enhanced by a curated large‑scale trajectory corpus and a reinforcement learning paradigm with multi‑turn refinement and turn‑level credit assignment, achieving superior zero‑shot forecasting performance.

By Kun Feng, Ziwei Shan, Yuchen Fang, Yiyang Tan, Sihan Lu, Shuqi Gu, Xingyu Lu, Lintao Ma, Kan Ren
arXiv AI
1d ago

TimesJev: Thinking Fast on Time Series with System 1 Sufficiency

TimesJev is a System 1 time‑series model that compresses raw series into a representation preserving key frequency patterns. It optimizes a dual‑objective loss combining sufficiency divergence and realizability divergence, then aligns the compressed representation with the model’s embedding space. Experiments on 13 datasets show TimesJev outperforms reasoning models by over 62 % while reducing inference time by 62.9 % on average.

By Chenghao Liu, Wei Fan, Jindong Han, Dongjie Wang, Aniket Mahanti, Pengyang Wang, Hao Liu
arXiv AI
Nov 5

TSFM Meets LLM: Context as Covariates

The paper introduces Context as Covariates (CoCo), a forecasting framework that combines Large Language Models (LLMs) and Time Series Foundation Models (TSFMs). It generates realistic multimodal datasets by annotating real-world time series with LLMs and aligns the two models through staged fine‑tuning: first using Group Relative Policy Optimization to improve the LLM’s covariate generation, then fine‑tuning the TSFM on those covariates. Experiments on synthetic and real benchmarks across finance, economics, weather, traffic, and security show that CoCo can outperform state‑of‑the‑art TSFMs and LLM forecasters by up to 12% MAE.

By Hemanth Ram, Sai Shankar Narasimhan, Shubhankar Agarwal, Sandeep Chinchali
arXiv AI
Aug 18

Adapting LLMs to Time Series Forecasting via Temporal Heterogeneity Modeling and Representation Alignment

arXiv:2508. 07195v2 Announce Type: replace-cross Abstract: Recent advances have demonstrated that Large Language Models (LLMs) can be effectively adapted for time series forecasting, revealing strong potential beyond natural language tasks.

By Yanru Sun, Emadeldeen Eldele, Zongxia Xie, Yucheng Wang, Wenzhe Niu, Qinghua Hu, Chee Keong Kwoh, Min Wu
arXiv Machine Learning
Oct 9

Lapras: Latent Reasoning for Time Series Language Models

Lapras is a post‑training framework for Time Series Language Models that replaces explicit Chain‑of‑Thought reasoning with continuous latent thoughts in a joint time‑series‑language space. By self‑distilling from a teacher that generates textual rationales, the student model aligns its hidden states at the answer stage, transferring reasoning ability while producing only the final answer in text. Experiments on four TSLM backbones and five benchmarks show Lapras boosts average F1 by up to 10.79% over explicit CoT and reduces token generation by 23.9×, while still enabling readable reasoning traces through standard decoding.

By Yuliang Chen, Yu Yvonne Wu, Patrick Langer, Arvind Pillai, Sudarshan Regmi, Martin Maritsch, Juncheng Liu, Robert Jakob, Thomas Kaar, Tess Z. Griffin, Lisa Marsch, Michael V. Heinz, Nicholas C. Jacobson, Andrew Campbell