Skip to content
AI News HubLIVE
Original source2 min read

TSFM Meets LLM: Context as Covariates

Summary

This paper proposes Context as Covariates (CoCo), a forecasting framework where an LLM distills textual reasoning into numerically grounded forecast and confidence covariates to guide a TSFM backbone. The approach first fine-tunes the LLM via GRPO with a reward tied to forecasting improvement, then fine-tunes the TSFM on the aligned LLM's covariates. It achieves up to 12% MAE improvement over state-of-the-art TSFMs and LLM forecasters like GPT-4o on real benchmarks spanning finance, economics, weather, traffic, and security.

SourcearXiv Machine LearningAuthor: Hemanth Ram, Sai Shankar Narasimhan, Shubhankar Agarwal, Sandeep Chinchali
TSFM Meets LLM: Context as Covariates
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 3 Nov 2026]

Title:TSFM Meets LLM: Context as Covariates

View a PDF of the paper titled TSFM Meets LLM: Context as Covariates, by Hemanth Ram and 3 other authors

View PDF HTML (experimental)

Abstract:Textual information is ubiquitous in real-world time series datasets, appearing as news, reports, and annotations that hold key predictive signals not deducible from numerical history alone. Large Language Models (LLMs) interpret text but lack numerical precision, while Time Series Foundation Models (TSFMs) are quantitatively accurate but cannot reason about the effects of text. Moreover, the lack of large-scale paired multimodal datasets limits progress in bridging this gap. In this work, we address this challenge by providing a scalable approach for generating realistic multimodal datasets by annotating real-world time series with LLMs. Building on this, we propose Context as Covariates (CoCo), a forecasting framework in which an LLM distills its textual reasoning into numerically grounded forecast and confidence covariates that guide a TSFM backbone, thereby exploiting the strengths of both LLMs and TSFMs. We align the two models in stages, first fine-tuning the LLM via Group Relative Policy Optimization (GRPO) with a reward tied to covariate-induced improvement in forecasting performance, and then fine-tuning the TSFM on covariates generated by the aligned LLM. Our experiments show that models trained on our synthetic multimodal corpus generalize to real benchmarks spanning finance, economics, weather, traffic, and security. Notably, CoCo with Qwen3-4B-Instruct-2507 as the LLM and Chronos2 as the TSFM achieves up to 12% MAE improvement over state-of-the-art TSFMs and LLM forecasters, such as GPT-4o. Code, checkpoints, and the evaluation datasets are publicly available.

Subjects:

Machine Learning (cs.LG); Artificial Intelligence (cs.AI)

Cite as: arXiv:2611.05515 [cs.LG]

(or arXiv:2611.05515v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2611.05515

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Sai Shankar Narasimhan [view email] [v1] Tue, 3 Nov 2026 19:13:19 UTC (7,188 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled TSFM Meets LLM: Context as Covariates, by Hemanth Ram and 3 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.LG

new | recent | 2026-11

Change to browse by:

cs cs.AI

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender (What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • CoCo framework converts LLM textual reasoning into numerical covariates to compensate for LLM's lack of numerical precision and TSFM's inability to reason about text.
  • Staged alignment: first fine-tune LLM with GRPO using a reward tied to covariate-induced forecasting improvement, then fine-tune TSFM on covariates from the aligned LLM.
  • With Qwen3-4B-Instruct-2507 as LLM and Chronos2 as TSFM, CoCo achieves up to 12% MAE improvement over state-of-the-art TSFMs and LLM forecasters such as GPT-4o.
  • Code, checkpoints, and evaluation datasets are publicly available.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models