Time-LLM: Time Series Forecasting by Reprogramming Large Language Models

计算机科学 杠杆(统计) 人工智能 背景(考古学) 机器学习 时间序列 系列(地层学) 语言模型 模式 地理 古生物学 社会科学 考古 社会学 生物
作者
Ming Jin,Shiyu Wang,Lintao Ma,Zhixuan Chu,James Y. Zhang,Xiaoming Shi,Pin‐Yu Chen,Yuxuan Liang,Yuan-Fang Li,Shirui Pan,Qingsong Wen
出处
期刊:Cornell University - arXiv 被引量:123
标识
DOI:10.48550/arxiv.2310.01728
摘要

Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
浮游应助科研通管家采纳,获得10
刚刚
刚刚
wxyshare应助科研通管家采纳,获得10
刚刚
刚刚
刚刚
852应助科研通管家采纳,获得10
刚刚
烟花应助科研通管家采纳,获得10
刚刚
共享精神应助科研通管家采纳,获得10
刚刚
wxyshare应助科研通管家采纳,获得10
刚刚
刚刚
慕青应助科研通管家采纳,获得10
刚刚
小蘑菇应助科研通管家采纳,获得10
刚刚
刚刚
1秒前
1秒前
1秒前
1秒前
1秒前
夏天发布了新的文献求助10
1秒前
1秒前
平常紫安完成签到 ,获得积分10
1秒前
1秒前
领导范儿应助ljz910005采纳,获得20
2秒前
2秒前
dhaoini完成签到,获得积分10
2秒前
3秒前
慕青应助郭勇慧采纳,获得10
3秒前
善学以致用应助Doc_d采纳,获得10
3秒前
蚊蚊爱读书应助马家辉采纳,获得10
3秒前
研友_Zrl2pL完成签到,获得积分20
4秒前
果实发布了新的文献求助10
4秒前
4秒前
粥粥完成签到 ,获得积分10
5秒前
坚强紫山发布了新的文献求助10
5秒前
FashionBoy应助zy0411采纳,获得10
5秒前
5秒前
丘比特应助Daisy采纳,获得10
6秒前
单薄凝冬发布了新的文献求助10
6秒前
英勇的竺发布了新的文献求助10
6秒前
ggst发布了新的文献求助10
6秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Iron toxicity and hematopoietic cell transplantation: do we understand why iron affects transplant outcome? 1500
List of 1,091 Public Pension Profiles by Region 1001
EEG in Childhood Epilepsy: Initial Presentation & Long-Term Follow-Up 500
Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences 500
On the application of advanced modeling tools to the SLB analysis in NuScale. Part I: TRACE/PARCS, TRACE/PANTHER and ATHLET/DYN3D 500
L-Arginine Encapsulated Mesoporous MCM-41 Nanoparticles: A Study on In Vitro Release as Well as Kinetics 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 纳米技术 计算机科学 内科学 化学工程 复合材料 物理化学 基因 遗传学 催化作用 冶金 量子力学 光电子学
热门帖子
关注 科研通微信公众号,转发送积分 5473503
求助须知:如何正确求助?哪些是违规求助? 4575665
关于积分的说明 14353545
捐赠科研通 4503157
什么是DOI,文献DOI怎么找? 2467534
邀请新用户注册赠送积分活动 1455373
关于科研通互助平台的介绍 1429357