MetaCogAgent: A Metacognitive Multi-Agent LLM Framework with Self-Aware Task Delegation
The article introduces MetaCogAgent, a multi-agent large language model framework designed to enhance task delegation through self-awareness. It incorporates a Metacognitive Self-Assessment Unit that allows agents to evaluate their own capabilities before undertaking tasks. Experimental results indicate that this framework significantly improves task accuracy while reducing API calls compared to existing methods.
- ▪MetaCogAgent utilizes a self-assessment mechanism to estimate task confidence.
- ▪The framework features an adaptive delegation protocol for routing tasks to suitable agents.
- ▪Experiments show that MetaCogAgent achieves 82.4% task accuracy, outperforming the best routing baseline by 8.7%.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.17292 (cs) [Submitted on 17 May 2026] Title:MetaCogAgent: A Metacognitive Multi-Agent LLM Framework with Self-Aware Task Delegation Authors:Chenyu Wang, Yang Shu View a PDF of the paper titled MetaCogAgent: A Metacognitive Multi-Agent LLM Framework with Self-Aware Task Delegation, by Chenyu Wang and 1 other authors View PDF HTML (experimental) Abstract:Multi-agent large language model (LLM) systems have shown promise for solving complex tasks through agent collaboration. However, existing frameworks assign tasks based on predefined roles without considering whether an agent can accurately assess its own competence boundaries, leading to overconfident execution on tasks beyond its expertise.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
