mlsanityData to 14 Oct 2026

United Minds or Isolated Agents? Exploring Coordination of LLMs under Cognitive Load Theory

arXiv 2506.06843 ↗First posted 7 Jun 2025cs.AIcited by 10

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

Abstract (arXiv)

Large Language Models (LLMs) exhibit a notable performance ceiling on complex, multi-faceted tasks. As practitioners increasingly rely on heavy context engineering -- curating intricate instructions, tool schemas, and multi-turn histories -- the processing demands often exceed the LLM's effective attention budget, leading to context rot. Drawing an analogy to Cognitive Load Theory (CLT) in cognitive science, we propose that this bottleneck is functionally analogous to the bounded working memory of the human mind. Rather than relying on heuristic prompt engineering, we use CLT as a principled design lens for LLM system design. To operationalize this insight, we introduce CoThinker, an instantiation of a CLT-driven multi-agent framework. CoThinker operationalizes CLT principles by distributing intrinsic cognitive load through agent specialization and managing transactional load via structured communication and a collective working memory. We empirically evaluate CoThinker on complex problem-solving tasks and fabricated high cognitive load scenarios. Our results are consistent with a CLT-informed account of multi-agent coordination: gains concentrate on reasoning-heavy tasks where cognitive load is high, while coordination overhead dominates on low-intrinsic-load tasks such as instruction-following -- a boundary predicted by the cognitive-load-profile view. Our analysis reveals characteristic interaction patterns that cast insights from collective cognition and load management into a principled approach to agent system design.

Key facts. Cited by 10 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 58 of its 108 references that are in the map. First posted on arXiv 7 Jun 2025.

Builds on

58 of 108 references.

Closest ancestors · 12

Same direction, newest first

  1. 2024Improving Multi-Agent Debate with Sparse Communication Topology17 Jun
  2. 2024Scaling Large Language Model-based Multi-Agent Collaboration11 Jun
  3. 2024Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?28 Feb
  4. 2024Large Language Model based Multi-Agents: A Survey of Progress and Challenges21 Jan
  5. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Security10 Jan
  6. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents18 Oct
  7. 2023Theory of Mind for Multi-Agent Collaboration via Large Language Models16 Oct
  8. 2023A Survey on Large Language Model based Autonomous Agents22 Aug
  9. 2023ProAgent: Building Proactive Cooperative Agents with Large Language Models22 Aug
  10. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors21 Aug
  11. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework1 Aug
  12. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 46

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  5. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  6. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  7. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  8. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
  9. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  10. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,186
  11. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
  12. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,807
34 more
  1. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  2. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
  3. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,336
  4. 2023A Survey on Evaluation of Large Language Modelscited by 1,242
  5. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,228
  6. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  7. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 1,011
  8. 2023ChatDev: Communicative Agents for Software Developmentcited by 919
  9. 2019Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Prunedcited by 856
  10. 2025Why Do Multi-Agent LLM Systems Fail?cited by 604
  11. 2023Understanding the Effects of RLHF on LLM Generalisation and Diversitycited by 396
  12. 2023Explainability for Large Language Models: A Surveycited by 333
  13. 2023ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMscited by 309
  14. 2023Preference Ranking Optimization for Human Alignmentcited by 285
  15. 2023The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learningcited by 257
  16. 2023Large Language Model Alignment: A Surveycited by 243
  17. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 240
  18. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 222
  19. 2024Many-Shot In-Context Learningcited by 206
  20. 2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 201
  21. 2023Simulating Opinion Dynamics with Networks of LLM-based Agentscited by 154
  22. 2024Same Task, More Tokens: the Impact of Input Length on the Reasoning Performance of Large Language Modelscited by 152
  23. 2024LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Playcited by 66
  24. 2024Beyond Demographics: Aligning Role-playing LLM-based Agents Using Human Belief Networkscited by 39
  25. 2024Exploring Prosocial Irrationality for LLM Agents: A Social Cognition Viewcited by 38
  26. 2024Unveiling the Generalization Power of Fine-Tuned Large Language Modelscited by 36
  27. 2024The broader spectrum of in-context learningcited by 35
  28. 2024From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiencycited by 29
  29. 2024From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Modelscited by 27
  30. 2025Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answercited by 27
  31. 2024Attention Entropy is a Key Factor: An Analysis of Parallel Context Encoding with Full-attention-based Pre-trained Language Modelscited by 26
  32. 2023Working Memory Capacity of ChatGPT: An Empirical Studycited by 23
  33. 2024Is In-Context Learning Sufficient for Instruction Following in LLMs?cited by 19
  34. 2025Language Model Uncertainty Quantification with Attention Chaincited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 7.

10 papers.

  1. 2025Cognitive Load Limits in Large Language Models: Benchmarking Multi-Hop Reasoningcited by 5
  2. 2026SkillMAS: Skill Co-Evolution with LLM-based Multi-Agent Systemcited by 5
  3. 2026Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systemscited by 4
  4. 2026RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policycited by 4
  5. 2025Love First, Know Later: Persona-Based Romantic Compatibility Through LLM Text World Enginescited by 3
  6. 2026The Side Effects of Being Smart: Safety Risks in MLLMs' Multi-Image Reasoningcited by 3
  7. 2025Reasoning Relay: Evaluating Stability and Interchangeability of Large Language Models in Mathematical Reasoningcited by 2
  8. 2026FlowMind: Execute-Summarize for Structured Workflow Generation from LLM Reasoningcited by 0
  9. 2026Multi-Modal Multi-Agent Robotic Cognitive Alignment enabled by Non-Invasive Consumer Brain Computer Interfaces: A Proof of Concept Explorationcited by 0
  10. 2026S3: Spectral Null-Space Swap Makes Reasoning Models Efficientcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Accuracy: A Cognitive Load Framework for Mapping the Capability Boundaries of Tool-use Agents
  2. 2025Towards Cognitive Synergy in LLM-Based Multi-Agent Systems: Integrating Theory of Mind and Critical Evaluation
  3. 2026Agentic Cognitive Depth: Operational Criteria for Evaluating LLM Agents
  4. 2026Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering
  5. 2025Procedural Memory Is Not All You Need: Bridging Cognitive Gaps in LLM-Based Agents
  6. 2026From Cognitive Architectures to Language Agents: A Mechanism-Level Review of Lineage, Convergence, and Migration Gaps
  7. 2025What Do LLM Agents Do When Left Alone? Evidence of Spontaneous Meta-Cognitive Patterns
  8. 2026Brain-Inspired Graph Multi-Agent Systems for LLM Reasoning