Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart Environments
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918333334421504 |
|---|---|
| author | Doctorarastoo, Maral Flanigan, Katherine A. Bergés, Mario McComb, Christopher |
| author_facet | Doctorarastoo, Maral Flanigan, Katherine A. Bergés, Mario McComb, Christopher |
| contents | Anticipating human activities and their durations is essential in applications such as smart-home automation, simulation-based architectural and urban design, activity-based transportation system simulation, and human-robot collaboration, where adaptive systems must respond to human activities. Existing data-driven agent-based models--from rule-based to deep learning--struggle in low-data environments, limiting their practicality. This paper investigates whether large language models, pre-trained on broad human knowledge, can fill this gap by reasoning about everyday activities from compact contextual cues. We adopt a retrieval-augmented prompting strategy that integrates four sources of context--temporal, spatial, behavioral history, and persona--and evaluate it on the CASAS Aruba smart-home dataset. The evaluation spans two complementary tasks: next-activity prediction with duration estimation, and multi-step daily sequence generation, each tested with various numbers of few-shot examples provided in the prompt. Analyzing few-shot effects reveals how much contextual supervision is sufficient to balance data efficiency and predictive accuracy, particularly in low-data environments. Results show that large language models exhibit strong inherent temporal understanding of human behavior: even in zero-shot settings, they produce coherent daily activity predictions, while adding one or two demonstrations further refines duration calibration and categorical accuracy. Beyond a few examples, performance saturates, indicating diminishing returns. Sequence-level evaluation confirms consistent temporal alignment across few-shot conditions. These findings suggest that pre-trained language models can serve as promising temporal reasoners, capturing both recurring routines and context-dependent behavioral variations, thereby strengthening the behavioral modules of agent-based models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_11176 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart Environments Doctorarastoo, Maral Flanigan, Katherine A. Bergés, Mario McComb, Christopher Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science Machine Learning I.2.4; I.2.8; I.2.6; I.6.5; H.3.3 Anticipating human activities and their durations is essential in applications such as smart-home automation, simulation-based architectural and urban design, activity-based transportation system simulation, and human-robot collaboration, where adaptive systems must respond to human activities. Existing data-driven agent-based models--from rule-based to deep learning--struggle in low-data environments, limiting their practicality. This paper investigates whether large language models, pre-trained on broad human knowledge, can fill this gap by reasoning about everyday activities from compact contextual cues. We adopt a retrieval-augmented prompting strategy that integrates four sources of context--temporal, spatial, behavioral history, and persona--and evaluate it on the CASAS Aruba smart-home dataset. The evaluation spans two complementary tasks: next-activity prediction with duration estimation, and multi-step daily sequence generation, each tested with various numbers of few-shot examples provided in the prompt. Analyzing few-shot effects reveals how much contextual supervision is sufficient to balance data efficiency and predictive accuracy, particularly in low-data environments. Results show that large language models exhibit strong inherent temporal understanding of human behavior: even in zero-shot settings, they produce coherent daily activity predictions, while adding one or two demonstrations further refines duration calibration and categorical accuracy. Beyond a few examples, performance saturates, indicating diminishing returns. Sequence-level evaluation confirms consistent temporal alignment across few-shot conditions. These findings suggest that pre-trained language models can serve as promising temporal reasoners, capturing both recurring routines and context-dependent behavioral variations, thereby strengthening the behavioral modules of agent-based models. |
| title | Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart Environments |
| topic | Computation and Language Artificial Intelligence Computational Engineering, Finance, and Science Machine Learning I.2.4; I.2.8; I.2.6; I.6.5; H.3.3 |
| url | https://arxiv.org/abs/2602.11176 |