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

• URL: http://arxiv.org/abs/2504.09772v1
• Date: Mon, 14 Apr 2025 00:27:45 GMT
• Title: Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning
• Authors: Can Jin, Hongwu Peng, Qixin Zhang, Yujin Tang, Dimitris N. Metaxas, Tong Che,
• Abstract summary: Multi-agent systems (MAS) built on large language models (LLMs) offer a promising path toward solving complex, real-world tasks.<n>Recent advancements in test-time scaling (TTS) have significantly improved single-agent performance on challenging reasoning tasks.<n>We introduce an adaptive multi-agent framework designed to enhance collaborative reasoning through both model-level training and system-level coordination.
• Score: 29.580108004844856
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent systems (MAS) built on large language models (LLMs) offer a promising path toward solving complex, real-world tasks that single-agent systems often struggle to manage. While recent advancements in test-time scaling (TTS) have significantly improved single-agent performance on challenging reasoning tasks, how to effectively scale collaboration and reasoning in MAS remains an open question. In this work, we introduce an adaptive multi-agent framework designed to enhance collaborative reasoning through both model-level training and system-level coordination. We construct M500, a high-quality dataset containing 500 multi-agent collaborative reasoning traces, and fine-tune Qwen2.5-32B-Instruct on this dataset to produce M1-32B, a model optimized for multi-agent collaboration. To further enable adaptive reasoning, we propose a novel CEO agent that dynamically manages the discussion process, guiding agent collaboration and adjusting reasoning depth for more effective problem-solving. Evaluated in an open-source MAS across a range of tasks-including general understanding, mathematical reasoning, and coding-our system significantly outperforms strong baselines. For instance, M1-32B achieves 12% improvement on GPQA-Diamond, 41% on AIME2024, and 10% on MBPP-Sanitized, matching the performance of state-of-the-art models like DeepSeek-R1 on some tasks. These results highlight the importance of both learned collaboration and adaptive coordination in scaling multi-agent reasoning. Code is available at https://github.com/jincan333/MAS-TTS

Related papers

• Collab: Controlled Decoding using Mixture of Agents for LLM Alignment [90.6117569025754]
  Reinforcement learning from human feedback has emerged as an effective technique to align Large Language models.<n>Controlled Decoding provides a mechanism for aligning a model at inference time without retraining.<n>We propose a mixture of agent-based decoding strategies leveraging the existing off-the-shelf aligned LLM policies.
  arXiv  Detail & Related papers  (2025-03-27T17:34:25Z)
• MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents [59.825725526176655]
  Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents.<n>Existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition.<n>We introduce MultiAgentBench, a benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios.
  arXiv  Detail & Related papers  (2025-03-03T05:18:50Z)
• LLM-Powered Decentralized Generative Agents with Adaptive Hierarchical Knowledge Graph for Cooperative Planning [12.996741471128539]
  Developing intelligent agents for long-term cooperation in dynamic open-world scenarios is a major challenge in multi-agent systems.<n>We propose Decentralized Adaptive Knowledge Graph Memory and Structured Communication System (DAMCS) in a novel Multi-agent Crafter environment.<n>Our generative agents, powered by Large Language Models (LLMs), are more scalable than traditional MARL agents by leveraging external knowledge and language for long-term planning and reasoning.
  arXiv  Detail & Related papers  (2025-02-08T05:26:02Z)
• Multi-Agent Sampling: Scaling Inference Compute for Data Synthesis with Tree Search-Based Agentic Collaboration [81.45763823762682]
  This work aims to bridge the gap by investigating the problem of data synthesis through multi-agent sampling.<n>We introduce Tree Search-based Orchestrated Agents(TOA), where the workflow evolves iteratively during the sequential sampling process.<n>Our experiments on alignment, machine translation, and mathematical reasoning demonstrate that multi-agent sampling significantly outperforms single-agent sampling as inference compute scales.
  arXiv  Detail & Related papers  (2024-12-22T15:16:44Z)
• Towards Effective GenAI Multi-Agent Collaboration: Design and Evaluation for Enterprise Applications [15.480315462362531]
  This report presents a comprehensive evaluation of coordination and routing capabilities in a novel multi-agent collaboration framework.<n>For coordination capabilities, we demonstrate the effectiveness of inter-agent communication and payload referencing mechanisms, achieving end-to-end goal success rates of 90%.<n>Our analysis yields several key findings: multi-agent collaboration enhances goal success rates by up to 70% compared to single-agent approaches in our benchmarks.
  arXiv  Detail & Related papers  (2024-12-06T22:14:17Z)
• MALT: Improving Reasoning with Multi-Agent LLM Training [66.9481561915524]
  MALT (Multi-Agent LLM Training) is a novel post-training strategy that divides the reasoning process into generation, verification, and refinement steps.<n>On MATH, GSM8K, and CSQA, MALT surpasses the same baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40% respectively.
  arXiv  Detail & Related papers  (2024-12-02T19:30:36Z)
• Agent-Oriented Planning in Multi-Agent Systems [54.429028104022066]
  We propose AOP, a novel framework for agent-oriented planning in multi-agent systems.<n>In this study, we identify three critical design principles of agent-oriented planning, including solvability, completeness, and non-redundancy.<n> Extensive experiments demonstrate the advancement of AOP in solving real-world problems compared to both single-agent systems and existing planning strategies for multi-agent systems.
  arXiv  Detail & Related papers  (2024-10-03T04:07:51Z)
• What are the Essential Factors in Crafting Effective Long Context Multi-Hop Instruction Datasets? Insights and Best Practices [91.71951459594074]
  Long language models (LLMs) with extended context windows have significantly improved tasks such as information extraction, question answering, and complex planning scenarios. Existing methods typically utilize the Self-Instruct framework to generate instruction tuning data for better long context capability improvement. We propose the Multi-agent Interactive Multi-hop Generation framework, incorporating a Quality Verification Agent, a Single-hop Question Generation Agent, a Multiple Question Sampling Strategy, and a Multi-hop Question Merger Agent. Our findings show that our synthetic high-quality long-context instruction data significantly enhances model performance, even surpassing models trained on larger amounts of human
  arXiv  Detail & Related papers  (2024-09-03T13:30:00Z)
• BattleAgentBench: A Benchmark for Evaluating Cooperation and Competition Capabilities of Language Models in Multi-Agent Systems [15.159418172629701]
  Large Language Models (LLMs) are becoming increasingly powerful and capable of handling complex tasks. Compared to single agents, multi-agent systems have higher requirements for the collaboration capabilities of language models. We propose a benchmark, called BattleAgentBench, which defines seven sub-stages of three varying difficulty levels.
  arXiv  Detail & Related papers  (2024-08-28T17:43:55Z)
• A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration [55.35849138235116]
  We propose automatically selecting a team of agents from candidates to collaborate in a dynamic communication structure toward different tasks and domains. Specifically, we build a framework named Dynamic LLM-Powered Agent Network ($textDyLAN$) for LLM-powered agent collaboration. We demonstrate that DyLAN outperforms strong baselines in code generation, decision-making, general reasoning, and arithmetic reasoning tasks with moderate computational cost.
  arXiv  Detail & Related papers  (2023-10-03T16:05:48Z)
• ProAgent: Building Proactive Cooperative Agents with Large Language Models [89.53040828210945]
  ProAgent is a novel framework that harnesses large language models to create proactive agents. ProAgent can analyze the present state, and infer the intentions of teammates from observations. ProAgent exhibits a high degree of modularity and interpretability, making it easily integrated into various coordination scenarios.
  arXiv  Detail & Related papers  (2023-08-22T10:36:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.