CLaSP: Learning Concepts for Time-Series Signals from Natural Language Supervision
This addresses a scalability and adaptability issue for data scientists in domains like industrial diagnostics, though it is incremental as it builds on existing contrastive learning and LLM methods.
The paper tackles the problem of retrieving time-series signals using natural language queries by introducing CLaSP, which employs contrastive learning and leverages large language models to eliminate the need for predefined synonym dictionaries, achieving high accuracy on the TRUCE and SUSHI datasets.
This paper presents CLaSP, a novel model for retrieving time-series signals using natural language queries that describe signal characteristics. The ability to search time-series signals based on descriptive queries is essential in domains such as industrial diagnostics, where data scientists often need to find signals with specific characteristics. However, existing methods rely on sketch-based inputs, predefined synonym dictionaries, or domain-specific manual designs, limiting their scalability and adaptability. CLaSP addresses these challenges by employing contrastive learning to map time-series signals to natural language descriptions. Unlike prior approaches, it eliminates the need for predefined synonym dictionaries and leverages the rich contextual knowledge of large language models (LLMs). Using the TRUCE and SUSHI datasets, which pair time-series signals with natural language descriptions, we demonstrate that CLaSP achieves high accuracy in retrieving a variety of time series patterns based on natural language queries.