Saved in:
Bibliographic Details
Main Authors: Chow, Winnie, Gardiner, Lauren, Hallgrímsson, Haraldur T., Xu, Maxwell A., Ren, Shirley You
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.11376
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929615627354112
author Chow, Winnie
Gardiner, Lauren
Hallgrímsson, Haraldur T.
Xu, Maxwell A.
Ren, Shirley You
author_facet Chow, Winnie
Gardiner, Lauren
Hallgrímsson, Haraldur T.
Xu, Maxwell A.
Ren, Shirley You
contents Multi-modal large language models (MLLMs) have enabled numerous advances in understanding and reasoning in domains like vision, but we have not yet seen this broad success for time-series. Although prior works on time-series MLLMs have shown promising performance in time-series forecasting, very few works show how an LLM could be used for time-series reasoning in natural language. We propose a novel multi-modal time-series LLM approach that learns generalizable information across various domains with powerful zero-shot performance. First, we train a lightweight time-series encoder on top of an LLM to directly extract time-series information. Then, we fine-tune our model with chain-of-thought augmented time-series tasks to encourage the model to generate reasoning paths. We show that our model learns a latent representation that reflects specific time-series features (e.g. slope, frequency), as well as outperforming GPT-4o on a set of zero-shot reasoning tasks on a variety of domains.
format Preprint
id arxiv_https___arxiv_org_abs_2409_11376
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards Time Series Reasoning with LLMs
Chow, Winnie
Gardiner, Lauren
Hallgrímsson, Haraldur T.
Xu, Maxwell A.
Ren, Shirley You
Machine Learning
Multi-modal large language models (MLLMs) have enabled numerous advances in understanding and reasoning in domains like vision, but we have not yet seen this broad success for time-series. Although prior works on time-series MLLMs have shown promising performance in time-series forecasting, very few works show how an LLM could be used for time-series reasoning in natural language. We propose a novel multi-modal time-series LLM approach that learns generalizable information across various domains with powerful zero-shot performance. First, we train a lightweight time-series encoder on top of an LLM to directly extract time-series information. Then, we fine-tune our model with chain-of-thought augmented time-series tasks to encourage the model to generate reasoning paths. We show that our model learns a latent representation that reflects specific time-series features (e.g. slope, frequency), as well as outperforming GPT-4o on a set of zero-shot reasoning tasks on a variety of domains.
title Towards Time Series Reasoning with LLMs
topic Machine Learning
url https://arxiv.org/abs/2409.11376