Game of LLMs: Discovering Structural Constructs in Activities using Large Language Models

TL;DR

This paper explores using large language models to identify structural building blocks in human activities within smart homes, aiming to improve recognition of varied and infrequent activities.

Contribution

It introduces a novel approach leveraging large language models to discover activity building blocks, enhancing activity recognition in smart home environments.

Findings

Identified structural constructs in human activities using LLMs.

Improved recognition of short-duration and infrequent activities.

Proposed a new activity recognition method based on building blocks.

Abstract

Human Activity Recognition is a time-series analysis problem. A popular analysis procedure used by the community assumes an optimal window length to design recognition pipelines. However, in the scenario of smart homes, where activities are of varying duration and frequency, the assumption of a constant sized window does not hold. Additionally, previous works have shown these activities to be made up of building blocks. We focus on identifying these underlying building blocks--structural constructs, with the use of large language models. Identifying these constructs can be beneficial especially in recognizing short-duration and infrequent activities. We also propose the development of an activity recognition procedure that uses these building blocks to model activities, thus helping the downstream task of activity monitoring in smart homes.