Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

time series modality switching

Time series modality switching refers to the capability of machine learning systems to translate, convert, or transition representations between sequential numerical time-series data and other distinct data modalities, such as natural language text, vision, or speech. Within multimodal architectures and foundation models, this process bridges the semantic and structural gap between continuous temporal signals and discrete symbolic tokens. By mapping temporal patterns into unified feature spaces, modality switching allows systems to interpret raw numerical sequences to produce textual descriptions, answers, or visual representations, as well as generate or forecast time series from natural language prompts and instructions.

1 item

Position: What Can Large Language Models Tell Us about Time Series Analysis

Position: What Can Large Language Models Tell Us about Time Series Analysis

Ming Jin, Yifan Zhang, Wei Chen, Kexin Zhang, Yuxuan Liang, Bin Yang, Jindong Wang, Shirui Pan, Qingsong Wen

OrganizationsChinese Academy of SciencesEast China Normal UniversityGriffith UniversityMicrosoftSquirrel AIThe Hong Kong University of Science and TechnologyZhejiang University

Why you should read this

Categorizes the emerging roles of large language models in time series analysis as data enhancers, predictors, and autonomous agents while identifying concrete integration strategies and open research opportunities for building universal time series intelligence.

Time series analysis is essential for comprehending the complexities inherent in various real-world systems and applications. Although large language models (LLMs) have recently made significant strides, the development of artificial general intelligence (AGI) equipped with time series analysis capabilities remains in its nascent phase. Most existing time series models heavily rely on domain knowledge and extensive model tuning, predominantly focusing on prediction tasks. In this paper, we argue that current LLMs have the potential to revolutionize time series analysis, thereby promoting efficient decision-making and advancing towards a more universal form of time series analytical intelligence. Such advancement could unlock a wide range of possibilities, including time series modality switching and question answering. We encourage researchers and practitioners to recognize the potential of LLMs in advancing time series analysis and emphasize the need for trust in these related efforts. Furthermore, we detail the seamless integration of time series analysis with existing LLM technologies and outline promising avenues for future research.

Added

2026-09-26