Authors: MM Karim, S Khan, DH Van, X Liu, C Wang, Q Qu
Year: 2025
Published in: Future Internet, 2025 - mdpi.com
Institution: Chinese Academy of Sciences, Zhejiang University, South-Central Minzu University
Research Area: Artificial Intelligence, Data Annotation, Multi-Agent Systems
Discipline: Artificial Intelligence
The paper reviews the role of AI agents powered by large language models in addressing challenges in data annotation, focusing on architectures, workflows, real-world applications, and future research directions for improving efficiency, scalability, transparency, and bias mitigation.
Methods: Comprehensive review and analysis of AI agent architectures, workflows, applications, and evaluation methods in data annotation across multiple industries.
Key Findings: Capabilities of LLM-driven agents in reasoning, adaptive learning, collaborative annotation, and their impact on quality assurance, cost, scalability, and bias mitigation.
Citations: 10
