mlsanityData to 7 Oct 2026

AgentSquare: Automatic LLM Agent Search in Modular Design Space

arXiv 2410.06153 ↗First posted 8 Oct 2024cs.CLcited by 141

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Recent advancements in Large Language Models (LLMs) have led to a rapid growth of agentic systems capable of handling a wide range of complex tasks. However, current research largely relies on manual, task-specific design, limiting their adaptability to novel tasks. In this paper, we introduce a new research problem: Modularized LLM Agent Search (MoLAS). We propose a modular design space that abstracts existing LLM agent designs into four fundamental modules with uniform IO interface: Planning, Reasoning, Tool Use, and Memory. Building on this design space, we present a novel LLM agent search framework called AgentSquare, which introduces two core mechanisms, i.e., module evolution and recombination, to efficiently search for optimized LLM agents. To further accelerate the process, we design a performance predictor that uses in-context surrogate models to skip unpromising agent designs. Extensive experiments across six benchmarks, covering the diverse scenarios of web, embodied, tool use and game applications, show that AgentSquare substantially outperforms hand-crafted agents, achieving an average performance gain of 17.2% against best-known human designs. Moreover, AgentSquare can generate interpretable design insights, enabling a deeper understanding of agentic architecture and its impact on task performance. We believe that the modular design space and AgentSquare search framework offer a platform for fully exploiting the potential of prior successful designs and consolidating the collective efforts of research community. Code repo is available at https://github.com/tsinghua-fib-lab/AgentSquare.

Key facts. Cited by 141 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 51 of its 71 references that are in the map. First posted on arXiv 8 Oct 2024.

Builds on

51 of 71 references.

Closest ancestors · 8

Same direction, newest first

  1. 2024Automated Design of Agentic Systems15 Aug
  2. 2024EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms20 Jun
  3. 2024AgentGym: Evolving Large Language Model-based Agents across Diverse Environments6 Jun
  4. 2024TravelPlanner: A Benchmark for Real-World Planning with Language Agents2 Feb
  5. 2024AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents24 Jan
  6. 2023Cognitive Architectures for Language Agents5 Sep
  7. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors21 Aug
  8. 2023Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration11 Jul

Other references · 43

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  5. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  6. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  7. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  8. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  9. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  10. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,302
  11. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  12. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,722
31 more
  1. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,197
  2. 2023HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Facecited by 1,173
  3. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,143
  4. 2022Inner Monologue: Embodied Reasoning through Planning with Language Modelscited by 1,057
  5. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,029
  6. 2023ChatDev: Communicative Agents for Software Developmentcited by 893
  7. 2023Reasoning with Language Model is Planning with World Modelcited by 846
  8. 2023Large Language Models as Optimizerscited by 830
  9. 2024Executable Code Actions Elicit Better LLM Agentscited by 537
  10. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 521
  11. 2022ScienceWorld: Is your Agent Smarter than a 5th Grader?cited by 298
  12. 2024Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agentscited by 210
  13. 2023DiLu: A Knowledge-Driven Approach to Autonomous Driving with Large Language Modelscited by 204
  14. 2023OpenAGI: When LLM Meets Domain Expertscited by 196
  15. 2019The Evolved Transformercited by 180
  16. 2019BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Searchcited by 175
  17. 2023Take a Step Back: Evoking Reasoning via Abstraction in Large Language Modelscited by 157
  18. 2022Evolution through Large Modelscited by 131
  19. 2023EvoPrompting: Language Models for Code-Level Neural Architecture Searchcited by 129
  20. 2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 86
  21. 2023MetaAgents: Large Language Model Based Agents for Decision-Making on Teamingcited by 85
  22. 2024AutoFlow: Automated Workflow Generation for Large Language Model Agentscited by 76
  23. 2019BayesNAS: A Bayesian Approach for Neural Architecture Searchcited by 67
  24. 2024Large Language Model Agent in Financial Trading: A Surveycited by 60
  25. 2023Thought Propagation: An Analogical Approach to Complex Reasoning with Large Language Modelscited by 37
  26. 2024Are Large Language Models Good Prompt Optimizers?cited by 37
  27. 2023LLM Performance Predictors are good initializers for Architecture Searchcited by 29
  28. 2024Perceive, Reflect, and Plan: Designing LLM Agent for Goal-Directed City Navigation without Instructionscited by 25
  29. 2024Large Language Model-driven Meta-structure Discovery in Heterogeneous Information Networkcited by 16
  30. 2024Automatic Prompt Selection for Large Language Modelscited by 9
  31. 2024LIMP: Large Language Model Enhanced Intent-aware Mobility Predictioncited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 51; 2026: 89.

141 papers.

Sort:

By area:

  1. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 291
  2. 2025Multi-agent Architecture Search via Agentic Supernetcited by 164
  3. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 110
  4. 2025Multi-Agent Design: Optimizing Agents with Better Prompts and Topologiescited by 106
  5. 2025Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memorycited by 88
  6. 2025AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systemscited by 86
  7. 2025MasRouter: Learning to Route LLMs for Multi-Agent Systemscited by 80
  8. 2024SPA-Bench: A Comprehensive Benchmark for SmartPhone Agent Evaluationcited by 79
  9. 2025Deep Research: A Survey of Autonomous Research Agentscited by 63
  10. 2025A Self-Improving Coding Agentcited by 58
  11. 2025FlowReasoner: Reinforcing Query-Level Meta-Agentscited by 56
  12. 2025FLEX: Continuous Agent Evolution via Forward Learning from Experiencecited by 51
38 more
  1. 2025EvoFlow: Evolving Diverse Agentic Workflows On The Flycited by 49
  2. 2025Adaptive Graph Pruning for Multi-Agent Communicationcited by 28
  3. 2025MAS-ZERO: Designing Multi-Agent Systems with Zero Supervisioncited by 26
  4. 2025Division-of-Thoughts: Harnessing Hybrid Language Model Synergy for Efficient On-Device Agentscited by 24
  5. 2025AgentRecBench: Benchmarking LLM Agent-based Personalized Recommender Systemscited by 23
  6. 2025Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systemscited by 22
  7. 2025DyFlow: Dynamic Workflow Framework for Agentic Reasoningcited by 22
  8. 2026ReCreate: Reasoning and Creating Domain Agents Driven by Experiencecited by 22
  9. 2026Graph-based Agent Memory: Taxonomy, Techniques, and Applicationscited by 22
  10. 2026From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flawscited by 21
  11. 2026HarnessForge: Joint Harness and Policy Evolution for Adaptive Agent Systemscited by 20
  12. 2025FingerTip 20K: A Benchmark for Proactive and Personalized Mobile LLM Agentscited by 19
  13. 2025RobustFlow: Towards Robust Agentic Workflow Generationcited by 19
  14. 2025Failure-Driven Workflow Refinementcited by 19
  15. 2025Learning to Deliberate: Meta-policy Collaboration for Agentic LLMs with Multi-agent Reinforcement Learningcited by 18
  16. 2025SwarmAgentic: Towards Fully Automated Agentic System Generation via Swarm Intelligencecited by 17
  17. 2026AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discoverycited by 17
  18. 2026Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judgecited by 16
  19. 2025AgentSwift: Efficient LLM Agent Design via Value-guided Hierarchical Searchcited by 14
  20. 2025Maestro: Learning to Collaborate via Conditional Listwise Policy Optimization for Multi-Agent LLMscited by 14
  21. 2025AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platformscited by 12
  22. 2025Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Modelscited by 11
  23. 2025Exploring Advanced LLM Multi-Agent Systems Based on Blackboard Architecturecited by 10
  24. 2025Alita-G: Self-Evolving Generative Agent for Agent Generationcited by 10
  25. 2026Mem2Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillationcited by 10
  26. 2026GraphPlanner: Graph Memory-Augmented Agentic Routing for Multi-Agent LLMscited by 10
  27. 2026JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolutioncited by 10
  28. 2025Aime: Towards Fully-Autonomous Multi-Agent Frameworkcited by 9
  29. 2026Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learningcited by 9
  30. 2026WHALE: A Simple Recipe for Joint Harness-Weight Optimizationcited by 9
  31. 2026The Interplay of Harness Design and Post-Training in LLM Agentscited by 8
  32. 2026Lifting Traces to Logic: Programmatic Skill Induction with Neuro-Symbolic Learning for Long-Horizon Agentic Taskscited by 7
  33. 2026Better Harnesses, Smaller Models: Building 90% Cheaper Agents via Automated Harness Adaptationcited by 7
  34. 2025Toward a Safe Internet of Agentscited by 6
  35. 2026Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priorscited by 6
  36. 2025AI Agent Behavioral Sciencecited by 5
  37. 2025Inefficiencies of Meta Agents for Agent Designcited by 5
  38. 2026Advancing Multimodal Agent Reasoning with Long-Term Neuro-Symbolic Memorycited by 5

50 most cited of 141 All 141 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024AgentQuest: A Modular Benchmark Framework to Measure Progress and Improve LLM Agents
  2. 2024Understanding the planning of LLM agents: A survey
  3. 2026Toward Formalizing LLM-Based Agent Designs through Structural Context Modeling and Semantic Dynamics Analysis
  4. 2025LLMs Working in Harmony: A Survey on the Technological Aspects of Building Effective LLM-Based Multi Agent Systems
  5. 2026LLM Agents Factory: Retrieval of Domain-Specific LLM Agents
  6. 2025Large Language Model Agent: A Survey on Methodology, Applications and Challenges
  7. 2023AgentBench: Evaluating LLMs as Agents
  8. 2024Designing Algorithms Empowered by Language Models: An Analytical Framework, Case Studies, and Insights