mlsanityData to 7 Oct 2026

Investigate-Consolidate-Exploit: A General Strategy for Inter-Task Agent Self-Evolution

arXiv 2401.13996 ↗First posted 25 Jan 2024cs.CL cs.AIcited by 28

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

Abstract (arXiv)

This paper introduces Investigate-Consolidate-Exploit (ICE), a novel strategy for enhancing the adaptability and flexibility of AI agents through inter-task self-evolution. Unlike existing methods focused on intra-task learning, ICE promotes the transfer of knowledge between tasks for genuine self-evolution, similar to human experience learning. The strategy dynamically investigates planning and execution trajectories, consolidates them into simplified workflows and pipelines, and exploits them for improved task execution. Our experiments on the XAgent framework demonstrate ICE's effectiveness, reducing API calls by as much as 80% and significantly decreasing the demand for the model's capability. Specifically, when combined with GPT-3.5, ICE's performance matches that of raw GPT-4 across various agent tasks. We argue that this self-evolution approach represents a paradigm shift in agent design, contributing to a more robust AI community and ecosystem, and moving a step closer to full autonomy.

Key facts. Cited by 28 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 31 of its 42 references that are in the map. First posted on arXiv 25 Jan 2024.

Builds on

31 of 42 references.

Closest ancestors · 7

Same direction, newest first

  1. 2023The Rise and Potential of Large Language Model Based Agents: A Survey14 Sep
  2. 2023Cognitive Architectures for Language Agents5 Sep
  3. 2023A Survey on Large Language Model based Autonomous Agents22 Aug
  4. 2023ExpeL: LLM Agents Are Experiential Learners20 Aug
  5. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate14 Aug
  6. 2023AgentBench: Evaluating LLMs as Agents7 Aug
  7. 2023AdaPlanner: Adaptive Planning from Feedback with Language Models26 May

Other references · 24

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  3. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  4. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  5. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  6. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  7. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  8. 2022Emergent Abilities of Large Language Modelscited by 2,483
  9. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  10. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
  11. 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,302
  12. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
12 more
  1. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  2. 2022GLM-130B: An Open Bilingual Pre-trained Modelcited by 837
  3. 2023Large Language Models as Optimizerscited by 830
  4. 2014Neural Turing Machinescited by 630
  5. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 628
  6. 2022LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Modelscited by 459
  7. 2023A Real-World WebAgent with Planning, Long Context Understanding, and Program Synthesiscited by 350
  8. 2023Large Language Models as Tool Makerscited by 293
  9. 2023Augmenting Language Models with Long-Term Memorycited by 186
  10. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoningcited by 157
  11. 2023Toolink: Linking Toolkit Creation and Using through Chain-of-Solving on Open-Source Modelcited by 29
  12. 2023Automatic Engineering of Long Promptscited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 9; 2025: 10; 2026: 9.

28 papers.

Sort:

By area:

  1. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 253
  2. 2026Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skillscited by 161
  3. 2024Symbolic Learning Enables Self-Evolving Agentscited by 104
  4. 2024A Review of Large Language Models and Autonomous Agents in Chemistrycited by 96
  5. 2024Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimizationcited by 93
  6. 2024A Survey on Self-Evolution of Large Language Modelscited by 86
  7. 2025Acting Less is Reasoning More! Teaching Model to Act Efficientlycited by 80
  8. 2025UserBench: An Interactive Gym Environment for User-Centric Agentscited by 69
  9. 2025A Survey on (M)LLM-Based GUI Agentscited by 61
  10. 2025SMART: Self-Aware Agent for Tool Overuse Mitigationcited by 52
  11. 2025ChemAgent: Self-updating Library in Large Language Models Improves Chemical Reasoningcited by 41
  12. 2024How Far Are We From AGI: Are LLMs All We Need?cited by 31
16 more
  1. 2024G"odel Agent: A Self-Referential Agent Framework for Recursive Self-Improvementcited by 27
  2. 2025Training LLM-Based Agents with Synthetic Self-Reflected Trajectories and Partial Maskingcited by 16
  3. 2024MobA: Multifaceted Memory-Enhanced Adaptive Planning for Efficient Mobile Task Automationcited by 14
  4. 2025Generalizable End-to-End Tool-Use RL with Synthetic CodeGymcited by 14
  5. 2024MSI-Agent: Incorporating Multi-Scale Insight into Embodied Agents for Superior Planning and Decision-Makingcited by 9
  6. 2024ReDel: A Toolkit for LLM-Powered Recursive Multi-Agent Systemscited by 8
  7. 2025ModelingAgent: Bridging LLMs and Mathematical Modeling for Real-World Challengescited by 4
  8. 2026CoMAP: Co-Evolving World Models and Agent Policies for LLM Agentscited by 4
  9. 2025How Far Can LLMs Improve from Experience? Measuring Test-Time Learning Ability in LLMs with Human Comparisoncited by 3
  10. 2026Self-evolving LLM agents with in-distribution Optimizationcited by 3
  11. 2026SKILL.nb: Selective Formalization and Gated Execution for Durable Agent Workflowscited by 3
  12. 2026AutoRPA: Efficient GUI Automation through LLM-Driven Code Synthesis from Interactionscited by 2
  13. 2026Online Skill Learning for Web Agents via State-Grounded Dynamic Retrievalcited by 2
  14. 2026Controlling Tool Use with Heading-Specific Activation Steeringcited by 2
  15. 2026When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoningcited by 1
  16. 2026OptiSkill: A Hierarchical and Evolving SkillBank for LLM-Based Optimization Modelingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SEA-Eval: A Benchmark for Evaluating Self-Evolving Agents Beyond Episodic Assessment
  2. 2026Self-Evolving Skills via Surrogate-Guided Solve-and-Reproduce
  3. 2026EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer
  4. 2025A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
  5. 2025Evolving in Tasks: Empowering the Multi-modality Large Language Model as the Computer Use Agent
  6. 2025Towards Self-Evolving Benchmarks: Synthesizing Agent Trajectories via Test-Time Exploration under Validate-by-Reproduce Paradigm
  7. 2026AutoAgent: Evolving Cognition and Elastic Memory Orchestration for Adaptive Agents
  8. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence