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

Authors: 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

ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 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.
Researcher Affiliation Collaboration 1Monash University 2Ant Group 3IBM Research 4Griffith University 5Alibaba Group 6The Hong Kong University of Science and Technology (Guangzhou)
Pseudocode No The paper describes the model structure in text and diagrams but does not provide any formal pseudocode or algorithm blocks.
Open Source Code Yes The code is made available at https://github.com/Kim Meen/Time-LLM.
Open Datasets Yes We evaluate on ETTh1, ETTh2, ETTm1, ETTm2, Weather, Electricity (ECL), Traffic, and ILI, which have been extensively adopted for benchmarking long-term forecasting models (Wu et al., 2023). ... Dataset statistics are summarized in Tab. 8.
Dataset Splits Yes Dataset statistics are summarized in Tab. 8. The dimension indicates the number of time series (i.e., channels), and the dataset size is organized in (training, validation, testing).
Hardware Specification Yes Our model implementation is on Py Torch (Paszke et al., 2019) with all experiments conducted on NVIDIA A100-80G GPUs.
Software Dependencies No The paper mentions 'Py Torch (Paszke et al., 2019)' but does not specify a version number for PyTorch itself, nor does it list other software components with specific version numbers.
Experiment Setup Yes The configurations of our models, relative to varied tasks and datasets, are consolidated in Tab. 9. ...Table 9: An overview of the experimental configurations for TIME-LLM. LTF and STF denote long-term and short-term forecasting, respectively.