# Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning

arXiv: https://arxiv.org/abs/2504.09772 · HTML: https://mlsanity.com/paper/2504.09772/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 84 of its 111 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- 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, 30 of 84)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society](https://mlsanity.com/paper/2303.17760/) (2023)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery](https://mlsanity.com/paper/2408.06292/) (2024)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [In-context Learning and Induction Heads](https://mlsanity.com/paper/2209.11895/) (2022)

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

- [ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory](https://mlsanity.com/paper/2509.25140/) (2025)
- [Thinking vs. Doing: Agents that Reason by Scaling Test-Time Interaction](https://mlsanity.com/paper/2506.07976/) (2025)
- [AgentTTS: Large Language Model Agent for Test-time Compute-optimal Scaling Strategy in Complex Tasks](https://mlsanity.com/paper/2508.00890/) (2025)
- [AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning](https://mlsanity.com/paper/2508.14313/) (2025)
- [How to Train a Leader: Hierarchical Reasoning in Multi-Agent LLMs](https://mlsanity.com/paper/2507.08960/) (2025)
- [MTL-UE: Learning to Learn Nothing for Multi-Task Learning](https://mlsanity.com/paper/2505.05279/) (2025)
- [Scaling over Scaling: Exploring Test-Time Scaling Plateau in Large Reasoning Models](https://mlsanity.com/paper/2505.20522/) (2025)
- [M^3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark](https://mlsanity.com/paper/2511.17729/) (2025)
- [Behavior Knowledge Merge in Reinforced Agentic Models](https://mlsanity.com/paper/2601.13572/) (2026)
- [Web Fraud Attacks Against LLM-Driven Multi-Agent Systems](https://mlsanity.com/paper/2509.01211/) (2025)
- [Revisiting Test-Time Scaling: A Survey and a Diversity-Aware Method for Efficient Reasoning](https://mlsanity.com/paper/2506.04611/) (2025)
- [Tool Verification for Test-Time Reinforcement Learning](https://mlsanity.com/paper/2603.02203/) (2026)
- [Agent Skill Evaluation and Evolution: Frameworks and Benchmarks](https://mlsanity.com/paper/2606.11435/) (2026)
- [MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems](https://mlsanity.com/paper/2602.03053/) (2026)
- [CoNav: Collaborative Cross-Modal Reasoning for Embodied Navigation](https://mlsanity.com/paper/2505.16663/) (2025)
- [A Theory of Inference Compute Scaling: Reasoning through Directed Stochastic Skill Search](https://mlsanity.com/paper/2507.00004/) (2025)
- [EMO-R3: Reflective Reinforcement Learning for Emotional Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2602.23802/) (2026)
- [LiteCUA: Computer as MCP Server for Computer-Use Agent on AIOS](https://mlsanity.com/paper/2505.18829/) (2025)
- [A Survey of Deep Learning for Geometry Problem Solving](https://mlsanity.com/paper/2507.11936/) (2025)
- [Language-Driven Coordination and Learning in Multi-Agent Simulation Environments](https://mlsanity.com/paper/2506.04251/) (2025)
- [Reasoning over Precedents Alongside Statutes: Case-Augmented Deliberative Alignment for LLM Safety](https://mlsanity.com/paper/2601.08000/) (2026)
- [MARTI-MARS2: Scaling Multi-Agent Self-Search via Reinforcement Learning for Code Generation](https://mlsanity.com/paper/2602.07848/) (2026)
- [Structured Debate Improves Corporate Credit Reasoning in Financial AI](https://mlsanity.com/paper/2510.17108/) (2025)
- [Explainable reinforcement learning from human feedback to improve alignment](https://mlsanity.com/paper/2512.13837/) (2025)
- [Scaling Unverifiable Rewards: A Case Study on Visual Insights](https://mlsanity.com/paper/2512.22650/) (2025)
- [E-bike agents: Large Language Model-Driven E-Bike Accident Analysis and Severity Prediction](https://mlsanity.com/paper/2506.04654/) (2025)
- [TokenBlowUp: Resolving Representational Singularities in LLM Token Spaces via Monoidal Transformations](https://mlsanity.com/paper/2507.19747/) (2025)
- [DIANOIA: Diagnostic Decomposition and Joint Optimization for Multi-Agent Reasoning](https://mlsanity.com/paper/2602.08586/) (2026)
- [Ecology of AI Agents: Collaboration Creates a Population Threshold for Takeoff](https://mlsanity.com/paper/2610.12436/) (2026)

## Abstract (arXiv)

Test-Time Scaling has emerged as a powerful method to extend the reasoning capabilities of Large Language Models. However, single-agent TTS faces significant scalability bottlenecks, as excessively long reasoning traces lead to increased inference costs and stability issues caused by context management failures. To address these limitations, we propose leveraging Multi-Agent Systems as a structural upgrade to standard TTS. By decomposing monolithic reasoning chains into distinct, manageable contexts across multiple agents, MAS offers a more robust framework for scaling reasoning. We validate this approach by introducing M500, a dataset comprising 500 high-quality multi-agent, multi-turn collaborative reasoning traces generated via DeepSeek-R1. Through Supervised Fine-Tuning on M500, we enable open-source models to internalize collaborative reasoning patterns and show improved TTS performance in MAS. Furthermore, we propose an adaptive scaling strategy incorporating a ``CEO'' agent to dynamically guide the reasoning process and optimize collaboration depth. Extensive experiments within the AgentVerse framework demonstrate that our fine-tuned models, Qwen2.5-32B-MAS and Phi4-14B-MAS, significantly outperform their base counterparts. Codes are available at https://github.com/jincan333/MAS-TTS.