Meta-reasoning in autonomous agents: performance gains across benchmarks and models
Academia AI and Applications · 2026
DOI: 10.20935/acadai8229 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academia AI and Applications, 2026. Verfasst von Wrick Talukdar und 4 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben.
Abstract (Originalsprache)
Introduction: Meta-reasoning, the ability of an autonomous agent to monitor and regulate its own reasoning processes, has emerged as a critical component for achieving adaptive and trustworthy artificial intelligence. However, limited empirical evidence quantifies how meta-reasoning impacts agent performance across diverse contexts and model architectures. This study presents a comprehensive empirical analysis of meta-reasoning capabilities in large language model (LLM)-based autonomous agents using two established benchmarks: GAIA (General AI Assistants) and AgentBench.
Zitieren
APA
Talukdar, W., Biswas, A., Shankar, G., Shinde, V., & Parekh, G. (2026). Meta-reasoning in autonomous agents: performance gains across benchmarks and models. Academia AI and Applications. https://doi.org/10.20935/acadai8229
Vancouver
Talukdar W, Biswas A, Shankar G, Shinde V, Parekh G. Meta-reasoning in autonomous agents: performance gains across benchmarks and models. Academia AI and Applications. 2026. doi:10.20935/acadai8229
BibTeX
@article{Talukdar2026,
author = {Talukdar, Wrick and Biswas, Anjanava and Shankar, Gowtham and Shinde, Varun and Parekh, Gaurav},
title = {Meta-reasoning in autonomous agents: performance gains across benchmarks and models},
journal = {Academia AI and Applications},
year = {2026},
doi = {10.20935/acadai8229}
}
EUVIMED