A Comprehensive Review of AI Agents: Transforming Possibilities in Technology and Beyond
On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks
Abstract (arXiv)
Artificial Intelligence (AI) agents have rapidly evolved from specialized, rule-based programs to versatile, learning-driven autonomous systems capable of perception, reasoning, and action in complex environments. The explosion of data, advances in deep learning, reinforcement learning, and multi-agent coordination have accelerated this transformation. Yet, designing and deploying unified AI agents that seamlessly integrate cognition, planning, and interaction remains a grand challenge. In this review, we systematically examine the architectural principles, foundational components, and emergent paradigms that define the landscape of contemporary AI agents. We synthesize insights from cognitive science-inspired models, hierarchical reinforcement learning frameworks, and large language model-based reasoning. Moreover, we discuss the pressing ethical, safety, and interpretability concerns associated with deploying these agents in real-world scenarios. By highlighting major breakthroughs, persistent challenges, and promising research directions, this review aims to guide the next generation of AI agent systems toward more robust, adaptable, and trustworthy autonomous intelligence.
Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 70 of its 163 references that are in the map. First posted on arXiv 16 Aug 2025.
Builds on
70 of 163 references.
Closest ancestors · 2
Same direction, newest first
- 2023The Rise and Potential of Large Language Model Based Agents: A Survey14 Sep
- 2023A Survey on Large Language Model based Autonomous Agents22 Aug
Other references · 68
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
- 2017Proximal Policy Optimization Algorithmscited by 15,215
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
- 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
- 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
- 2019Energy and Policy Considerations for Deep Learning in NLPcited by 1,254
38 more
- 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
- 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,071
- 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,035
- 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 634
- 2023API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMscited by 591
- 2023Augmented Language Models: a Surveycited by 403
- 2016Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivationcited by 286
- 2016A Survey of Motion Planning and Control Techniques for Self-driving Urban Vehiclescited by 254
- 2020Experience Grounds Languagecited by 242
- 2018Mean Field Multi-Agent Reinforcement Learningcited by 207
- 2020The Next Decade in AI: Four Steps Towards Robust Artificial Intelligencecited by 195
- 2020Emergent Multi-Agent Communication in the Deep Learning Eracited by 170
- 2024LAB-Bench: Measuring Capabilities of Language Models for Biology Researchcited by 156
- 2023Chain of Hindsight Aligns Language Models with Feedbackcited by 148
- 2019The Ethics of AI Ethics -- An Evaluation of Guidelinescited by 129
- 2024Empowering Biomedical Discovery with AI Agentscited by 118
- 2023Evaluating the Social Impact of Generative AI Systems in Systems and Societycited by 92
- 2016Towards an integration of deep learning and neurosciencecited by 78
- 2024ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLMcited by 73
- 2024Scaling Instructable Agents Across Many Simulated Worldscited by 71
- 2024MLR-Copilot: Autonomous Machine Learning Research based on Large Language Models Agentscited by 67
- 2022Few Could Be Better Than All: Feature Sampling and Grouping for Scene Text Detectioncited by 65
- 2024TextSquare: Scaling up Text-Centric Visual Instruction Tuningcited by 62
- 2024Harmonizing Visual Text Comprehension and Generationcited by 54
- 2024Large Language Models for Forecasting and Anomaly Detection: A Systematic Literature Reviewcited by 53
- 2024Exploring Diverse Methods in Visual Question Answeringcited by 45
- 2024ParGo: Bridging Vision-Language with Partial and Global Viewscited by 40
- 2022Optimal Boxes: Boosting End-to-End Scene Text Recognition by Adjusting Annotated Bounding Boxes via Reinforcement Learningcited by 33
- 2024Political-LLM: Large Language Models in Political Sciencecited by 27
- 2021ReLAX: Reinforcement Learning Agent eXplainer for Arbitrary Predictive Modelscited by 25
- 2022Incremental Learning Meets Transfer Learning: Application to Multi-site Prostate MRI Segmentationcited by 22
- 2023Estimate-Then-Optimize versus Integrated-Estimation-Optimization versus Sample Average Approximation: A Stochastic Dominance Perspectivecited by 20
- 2024Enhance Image-to-Image Generation with LLaVA-generated Promptscited by 20
- 2024Advances in APPFL: A Comprehensive and Extensible Federated Learning Frameworkcited by 19
- 2023DesignGPT: Multi-Agent Collaboration in Designcited by 17
- 2025CT-PatchTST: Channel-Time Patch Time-Series Transformer for Long-Term Renewable Energy Forecastingcited by 16
- 2018Learning Quickly to Plan Quickly Using Modular Meta-Learningcited by 15
- 2024From MOOC to MAIC: Reshaping Online Teaching and Learning through LLM-driven Agentscited by 15
50 most cited of 68 All 68 (JSON)
Built on by
Citing papers per year: 2025: 3; 2026: 3.
6 papers.
- 2026InfiAgent: An Infinite-Horizon Framework for General-Purpose Autonomous Agentscited by 10
- 2025InfiAgent: Self-Evolving Pyramid Agent Framework for Infinite Scenarioscited by 4
- 2025Reliable agent engineering should integrate machine-compatible organizational principlescited by 2
- 2025Modeling User-System Behavior for Training-free Building of Private Domain Conversational Agentscited by 0
- 2026Can Induced Emotion Bias LLM Behaviors in Sequential Decision Making?cited by 0
- 2026Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation taskscited by 0
- 2026Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks23 Jul 2026
- 2026Can Induced Emotion Bias LLM Behaviors in Sequential Decision Making?14 Jul 2026
- 2026InfiAgent: An Infinite-Horizon Framework for General-Purpose Autonomous Agents6 Jan 2026
- 2025Reliable agent engineering should integrate machine-compatible organizational principles8 Dec 2025
- 2025Modeling User-System Behavior for Training-free Building of Private Domain Conversational Agents13 Nov 2025
- 2025InfiAgent: Self-Evolving Pyramid Agent Framework for Infinite Scenarios26 Sep 2025
Similar, not cited
Close in wording, no citation either way
- 2025AI Agents: Evolution, Architecture, and Real-World Applications
- 2026AI Agent Systems: Architectures, Applications, and Evaluation
- 2024The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey
- 2026Agentic Artificial Intelligence (AI): Architectures, Taxonomies, and Evaluation of Large Language Model Agents
- 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems
- 2025AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges
- 2026The Path Ahead for Agentic AI: Challenges and Opportunities
- 2025Distinguishing Autonomous AI Agents from Collaborative Agentic Systems: A Comprehensive Framework for Understanding Modern Intelligent Architectures