Abstract
This research delves into the capabilities of Large Language Model (LLM)-powered agents across five critical dimensions of task management: decomposition, scheduling, delegation, and execution. Leveraging both surveys and interaction experiments, this study aims to unearth practical insights into the applications and constraints of LLM-powered agents in real-world settings. We describe our experimental setup, data collection methodologies, and analytic techniques, offering a nuanced understanding of these agents' efficiency and efficacy. Our initial findings underscore the nuanced performance of LLM-powered agents in task management, revealing their strengths in understanding complex tasks and their limitations in execution without human intervention. This research contributes to the burgeoning field of human-AI collaboration by providing empirical evidence on the capabilities and limitations of LLM-powered agents in task management.
Cite this work
Ravindu Tharanga Perera, Adithya Ravi, and Carlos Toxtli-Hernández. 2024. Assessing the Task Management Capabilities of LLM-Powered Agents. CHIWORK '24: 3rd Annual Meeting of the Symposium on Human-Computer Interaction for Work, Demo Track. https://doi.org/10.13140/RG.2.2.11776.85768
@misc{Perera2024Assessing,
doi = {10.13140/RG.2.2.11776.85768},
url = {https://www.researchgate.net/doi/10.13140/RG.2.2.11776.85768},
author = {{Ravindu Perera} and {Adithya Ravi} and Toxtli, Carlos},
language = {en},
title = {Assessing the Task Management Capabilities of LLM-Powered Agents},
publisher = {Unpublished},
year = {2024},
howpublished = {Preprint, ResearchGate},
note = {Preprint}
}
