# Understanding Agent Scaling in LLM-Based Multi-Agent Systems via Diversity

arXiv: https://arxiv.org/abs/2602.03794 · HTML: https://mlsanity.com/paper/2602.03794/

## Key facts

- Cited by 29 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 25 of its 44 references that are in the map.
- First posted on arXiv 3 Feb 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 25 of 25)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [The Rise and Potential of Large Language Model Based Agents: A Survey](https://mlsanity.com/paper/2309.07864/) (2023)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [Large Language Model based Multi-Agents: A Survey of Progress and Challenges](https://mlsanity.com/paper/2402.01680/) (2024)
- [Mixture-of-Agents Enhances Large Language Model Capabilities](https://mlsanity.com/paper/2406.04692/) (2024)
- [More Agents Is All You Need](https://mlsanity.com/paper/2402.05120/) (2024)
- [Towards a Science of Scaling Agent Systems](https://mlsanity.com/paper/2512.08296/) (2025)
- [PersonaGym: Evaluating Persona Agents and LLMs](https://mlsanity.com/paper/2407.18416/) (2024)
- [Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?](https://mlsanity.com/paper/2508.17536/) (2025)
- [Understanding Chain-of-Thought in LLMs through Information Theory](https://mlsanity.com/paper/2411.11984/) (2024)
- [Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents](https://mlsanity.com/paper/2408.07060/) (2024)
- [Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning](https://mlsanity.com/paper/2501.15602/) (2025)
- [Are More LLM Calls All You Need? Towards Scaling Laws of Compound Inference Systems](https://mlsanity.com/paper/2403.02419/) (2024)
- [Emergent Coordination in Multi-Agent Language Models](https://mlsanity.com/paper/2510.05174/) (2025)
- [Intrinsic Memory Agents: Heterogeneous Multi-Agent LLM Systems through Structured Contextual Memory](https://mlsanity.com/paper/2508.08997/) (2025)
- [The Hidden Strength of Disagreement: Unraveling the Consensus-Diversity Tradeoff in Adaptive Multi-Agent Systems](https://mlsanity.com/paper/2502.16565/) (2025)
- [From the Fluency Fallacy to the Micro-to-Macro Validity Gap: Opportunities and Pitfalls of LLMs in Social Simulation](https://mlsanity.com/paper/2507.19364/) (2025)
- [On the Importance of Task Complexity in Evaluating LLM-Based Multi-Agent Systems](https://mlsanity.com/paper/2510.04311/) (2025)

## Built on by (most cited first, 29 of 29)

- [Effective Strategies for Asynchronous Software Engineering Agents](https://mlsanity.com/paper/2603.21489/) (2026)
- [AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation](https://mlsanity.com/paper/2605.28655/) (2026)
- [From Model Scaling to System Scaling: Scaling the Harness in Agentic AI](https://mlsanity.com/paper/2605.26112/) (2026)
- [Epistemic Gain, Aleatoric Cost: Uncertainty Decomposition in Multi-Agent Debate for Math Reasoning](https://mlsanity.com/paper/2603.01221/) (2026)
- [SkillMAS: Skill Co-Evolution with LLM-based Multi-Agent System](https://mlsanity.com/paper/2605.09341/) (2026)
- [MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?](https://mlsanity.com/paper/2606.23664/) (2026)
- [Social Simulacra in the Wild: AI Agent Communities on Moltbook](https://mlsanity.com/paper/2603.16128/) (2026)
- [The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size](https://mlsanity.com/paper/2606.02646/) (2026)
- [What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents](https://mlsanity.com/paper/2608.27260/) (2026)
- [Improving the Efficiency of Language Agent Teams with Adaptive Task Graphs](https://mlsanity.com/paper/2605.06320/) (2026)
- [Oracle Gap and Signal Fidelity: A Fixed-Pool Diagnostic for Test-Time Collaboration](https://mlsanity.com/paper/2607.17531/) (2026)
- [MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts](https://mlsanity.com/paper/2608.09251/) (2026)
- [MOTOR-Bench: A Real-world Dataset and Multi-agent Framework for Zero-shot Human Mental State Understanding](https://mlsanity.com/paper/2605.09703/) (2026)
- [When Does Multi-Agent RL Improve LLM Workflows? Workflow, Scale, and Policy-Sharing Tradeoffs](https://mlsanity.com/paper/2605.24202/) (2026)
- [Spawn Freely, Act Sparingly: Progressive Risk Vesting for Recursive LLM-Agent Trees](https://mlsanity.com/paper/2609.01035/) (2026)
- [Beyond Alignment: Value Diversity as a Collective Property in Multicultural Agent Systems](https://mlsanity.com/paper/2606.05985/) (2026)
- [BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems](https://mlsanity.com/paper/2607.01600/) (2026)
- [Can We Steer the Black-Box? Towards Controllability-Centric Evaluation of Recommender Systems with Collaborative Agents](https://mlsanity.com/paper/2607.13418/) (2026)
- [Group Perspective Matters: Regulating Debate Relationships Can Mitigate Blind Conformity in Multi-Agent Debate](https://mlsanity.com/paper/2608.03648/) (2026)
- [A Structured Debate-Mixture-of-Agents Framework for Complex Clinical Diagnostic Decision Support](https://mlsanity.com/paper/2609.05069/) (2026)
- [WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies](https://mlsanity.com/paper/2609.05591/) (2026)
- [Mo' Models, Mo' Problems: How to best select model pools when designing Multi-Agent Systems](https://mlsanity.com/paper/2609.17306/) (2026)
- [Multi-agent Scaling Across Disjunctive and Compensatory Tasks](https://mlsanity.com/paper/2609.31563/) (2026)
- [RepoMAS: Solving Progressively Specified Tasks with Issue-Driven Multi-Agent Systems](https://mlsanity.com/paper/2609.32490/) (2026)
- [Towards Better Exploration in Sequential Test-Time Scaling](https://mlsanity.com/paper/2609.39632/) (2026)
- [Peer Influence across Heterogeneous AI Models](https://mlsanity.com/paper/2610.03095/) (2026)
- [Attention Tax, Handoff Tax: A Stylised Model of When Multi-Agent LLM Systems Help](https://mlsanity.com/paper/2610.06069/) (2026)
- [MAS-Gate: Unlocking Multi-Agent System Lifecycle via Node-Centric Quantification](https://mlsanity.com/paper/2610.17613/) (2026)
- [Scaling Discovery Requires Forgetting](https://mlsanity.com/paper/2610.20563/) (2026)

## Abstract (arXiv)

LLM-based multi-agent systems (MAS) have emerged as a promising approach to tackle complex tasks that are difficult for individual LLMs. A natural strategy is to scale performance by increasing the number of agents; however, we find that such scaling exhibits strong diminishing returns in homogeneous settings, while introducing heterogeneity (e.g., different models, prompts, or tools) continues to yield substantial gains. This raises a fundamental question: what limits scaling, and why does diversity help? We present an information-theoretic framework showing that MAS performance is bounded by the intrinsic task uncertainty, not by agent count. We derive architecture-agnostic bounds demonstrating that improvements depend on how many effective channels the system accesses. Homogeneous agents saturate early because their outputs are strongly correlated, whereas heterogeneous agents contribute complementary evidence. We further introduce K^*, an effective channel count that quantifies the number of effective channels without ground-truth labels. Empirically, we show that heterogeneous configurations consistently outperform homogeneous scaling: 2 diverse agents can match or exceed the performance of 16 homogeneous agents. Our results provide principled guidelines for building efficient and robust MAS through diversity-aware design. Code and Dataset are available at the link: https://github.com/SafeRL-Lab/Agent-Scaling.