LLM-Guided Exemplar Selection for Few-Shot Wearable-Sensor Human Activity Recognition

TL;DR

This paper introduces an LLM-guided exemplar selection framework that improves few-shot human activity recognition from wearable sensors by integrating semantic reasoning with structural cues, leading to more accurate activity classification.

Contribution

The novel framework combines LLM-derived semantic priors with geometric and structural cues for exemplar selection in few-shot HAR, outperforming classical methods.

Findings

Achieved 88.78% macro F1-score on UCI-HAR dataset.

Outperformed random sampling, herding, and k-center approaches.

Demonstrated the effectiveness of semantic priors in distinguishing similar activities.

Abstract

In this paper, we propose an LLM-Guided Exemplar Selection framework to address a key limitation in state-of-the-art Human Activity Recognition (HAR) methods: their reliance on large labeled datasets and purely geometric exemplar selection, which often fail to distinguish similar wearable sensor activities such as walking, walking upstairs, and walking downstairs. Our method incorporates semantic reasoning via an LLM-generated knowledge prior that captures feature importance, inter-class confusability, and exemplar budget multipliers, and uses it to guide exemplar scoring and selection. These priors are combined with margin-based validation cues, PageRank centrality, hubness penalization, and facility-location optimization to obtain a compact and informative set of exemplars. Evaluated on the UCI-HAR dataset under strict few-shot conditions, the framework achieves a macro F1-score of 88.78%, outperforming classical approaches such as random sampling, herding, and k-center. The results show that LLM-derived semantic priors, when integrated with structural and geometric cues, provide a stronger foundation for selecting representative sensor exemplars in few-shot wearable-sensor HAR.