ACL Findings2026
Multi-LLM Collaborative Search for Complex Problem Solving
Sen Yang, Yafu Li, Wai Lam, Yu Cheng
Abstract
Large language models (LLMs) often struggle with complex reasoning tasks due to their limitations in addressing the vast reasoning space and inherent ambiguities of natural language. We propose the Mixture-of-Search-Agents (MOSA) paradigm, a novel approach leveraging the collective expertise of multiple LLMs to enhance search-based reasoning. MOSA integrates diverse reasoning pathways by combining independent exploration with iterative refinement among LLMs, mitigating the limitations of single-model approaches. Using Monte Carlo Tree Search (MCTS) as a backbone, MOSA enables multiple agents to propose and aggregate reasoning steps, resulting in improved accuracy. Our comprehensive evaluation across four reasoning benchmarks demonstrates MOSA's consistent performance improvements over single-agent and other multiagent baselines, particularly in complex mathematical and commonsense reasoning tasks.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
4 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Let's Verify Step by Step cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Improving Factuality and Reasoning in Language Models through Multiagent Debate cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and Inference cites Let's Verify Step by Step
- MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and Inference cites Improving Factuality and Reasoning in Language Models through Multiagent Debate
- MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and Inference cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Cited by top-tier papers4
- Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMsYifan Shen, Yuanzhe Liu, Jingyuan Zhu, Xu Cao et al.NeurIPS 2025 · 41 citations
- MCCE: A Framework for Multi-LLM Collaborative Search in Discrete Spaces with Similarity-Filtered Preference LearningNian Ran, Zhongzheng Li, Yue Wang, Qingsong Ran et al.ICML 2026
- Diagnosing and Remedying Representation Deficiencies for Deterministic Reasoning in KGQAGewen Liang, Mufan Xu, Kehai Chen, Wei Wang et al.ACL 2026
- MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and InferenceKaiyan Zhang, Kai Tian, Runze Liu, Sihang Zeng et al.ICLR 2026
Builds on19
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran et al.NeurIPS 2023 · 5,068 citations
- The Curious Case of Neural Text DegenerationAri Holtzman, Jan Buys, Li Du, Maxwell Forbes et al.ICLR 2020 · 4,112 citations
- Let's Verify Step by StepHunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards et al.ICLR 2024 · 3,045 citations
- Improving Factuality and Reasoning in Language Models through Multiagent DebateYilun Du, Shuang Li, Antonio Torralba, Joshua B. Tenenbaum et al.ICML 2024 · 1,562 citations
Related papers
- DSG-MCTS: A Dynamic Strategy-Guided Monte Carlo Tree Search for Diversified Reasoning in Large Language ModelsRui Ha, Chaozhuo Li, Rui Pu, Litian Zhang et al.EMNLP 2025
- Balancing Act: Diversity and Consistency in Large Language Model EnsemblesAhmed Abdulaal, Chen Jin, Nina Montaña Brown, Aryo Pradipta Gema et al.ICLR 2025
- SYMPHONY: Synergistic Multi-agent Planning with Heterogeneous Language Model AssemblyWei Zhu, Zhiwen Tang, Kun YueNeurIPS 2025 · 10 citations
- Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table UnderstandingYuhang Zhou, Mingrui Zhang, Ke Li, Mingyi Wang et al.ACL 2026 · 5 citations
- Mixture-of-Agents Enhances Large Language Model CapabilitiesJunlin Wang, Jue Wang, Ben Athiwaratkun, Ce Zhang et al.ICLR 2025
