MARTI-MARS$^2$: Scaling Multi-Agent Self-Search via Reinforcement Learning for Code Generation

• URL: http://arxiv.org/abs/2602.07848v1
• Date: Sun, 08 Feb 2026 07:28:44 GMT
• Title: MARTI-MARS$^2$: Scaling Multi-Agent Self-Search via Reinforcement Learning for Code Generation
• Authors: Shijie Wang, Pengfei Li, Yikun Fu, Kaifeng Liu, Fangyuan Li, Yang Liu, Xiaowei Sun, Zonglin Li, Siyao Zhao, Jian Zhao, Kai Tian, Dong Li, Junqi Gao, Yutong Zhang, Yiqun Chen, Yuqiang Li, Zoe Li, Weinan Zhang, Peng Ye, Shuyue Hu, Lei Bai, Bowen Zhou, Kaiyan Zhang, Biqing Qi,
• Abstract summary: Multi-Agent Reinforced Training and Inference Framework with Self-Search Scaling (MARTI-MARS2)<n>We propose a Multi-Agent Reinforced Training and Inference Framework with Self-Search Scaling (MARTI-MARS2) to integrate policy learning with multi-agent tree search.<n>We show that MARTI-MARS2 achieves 77.7%, outperforming strong baselines like GPT-5.1 on challenging code generation benchmarks.
• Score: 64.2621682259008
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While the complex reasoning capability of Large Language Models (LLMs) has attracted significant attention, single-agent systems often encounter inherent performance ceilings in complex tasks such as code generation. Multi-agent collaboration offers a promising avenue to transcend these boundaries. However, existing frameworks typically rely on prompt-based test-time interactions or multi-role configurations trained with homogeneous parameters, limiting error correction capabilities and strategic diversity. In this paper, we propose a Multi-Agent Reinforced Training and Inference Framework with Self-Search Scaling (MARTI-MARS2), which integrates policy learning with multi-agent tree search by formulating the multi-agent collaborative exploration process as a dynamic and learnable environment. By allowing agents to iteratively explore and refine within the environment, the framework facilitates evolution from parameter-sharing homogeneous multi-role training to heterogeneous multi-agent training, breaking through single-agent capability limits. We also introduce an efficient inference strategy MARTI-MARS2-T+ to fully exploit the scaling potential of multi-agent collaboration at test time. We conduct extensive experiments across varied model scales (8B, 14B, and 32B) on challenging code generation benchmarks. Utilizing two collaborating 32B models, MARTI-MARS2 achieves 77.7%, outperforming strong baselines like GPT-5.1. Furthermore, MARTI-MARS2 reveals a novel scaling law: shifting from single-agent to homogeneous multi-role and ultimately to heterogeneous multi-agent paradigms progressively yields higher RL performance ceilings, robust TTS capabilities, and greater policy diversity, suggesting that policy diversity is critical for scaling intelligence via multi-agent reinforcement learning.

Related papers

• Collaborative Multi-Agent Test-Time Reinforcement Learning for Reasoning [112.16686518063456]
  We introduce textbfMulti-Agent Test-Time Reinforcement Learning (MATTRL), a framework that injects structured textual experience into multi-agent deliberation at inference time.<n>MATTRL forms a multi-expert team of specialists for multi-turn discussions, retrieves and integrates test-time experiences, and reaches consensus for final decision-making.<n>Across challenging benchmarks in medicine, math, and education, MATTRL improves accuracy by an average of 3.67% over a multi-agent baseline, and by 8.67% over comparable single-agent baselines
  arXiv  Detail & Related papers  (2026-01-14T17:57:43Z)
• AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning [129.44038804430542]
  We introduce AgentGym-RL, a new framework to train LLM agents for multi-turn interactive decision-making through RL.<n>We propose ScalingInter-RL, a training approach designed for exploration-exploitation balance and stable RL optimization.<n>Our agents match or surpass commercial models on 27 tasks across diverse environments.
  arXiv  Detail & Related papers  (2025-09-10T16:46:11Z)
• MisoDICE: Multi-Agent Imitation from Unlabeled Mixed-Quality Demonstrations [5.4482836906033585]
  We study offline imitation learning (IL) in cooperative multi-agent settings, where demonstrations have unlabeled mixed quality.<n>Our proposed solution is structured in two stages: trajectory labeling and multi-agent imitation learning.<n>We introduce MisoDICE, a novel multi-agent IL algorithm that leverages these labels to learn robust policies.
  arXiv  Detail & Related papers  (2025-05-24T08:43:42Z)
• Learning Emergence of Interaction Patterns across Independent RL Agents in Multi-Agent Environments [3.0284592792243794]
  Bottom Up Network (BUN) treats the collective of multi-agents as a unified entity. Our empirical evaluations across a variety of cooperative multi-agent scenarios, including tasks such as cooperative navigation and traffic control, consistently demonstrate BUN's superiority over baseline methods with substantially reduced computational costs.
  arXiv  Detail & Related papers  (2024-10-03T14:25:02Z)
• Multi-agent Deep Covering Skill Discovery [50.812414209206054]
  We propose Multi-agent Deep Covering Option Discovery, which constructs the multi-agent options through minimizing the expected cover time of the multiple agents' joint state space. Also, we propose a novel framework to adopt the multi-agent options in the MARL process. We show that the proposed algorithm can effectively capture the agent interactions with the attention mechanism, successfully identify multi-agent options, and significantly outperforms prior works using single-agent options or no options.
  arXiv  Detail & Related papers  (2022-10-07T00:40:59Z)
• Policy Diagnosis via Measuring Role Diversity in Cooperative Multi-agent RL [107.58821842920393]
  We quantify the agent's behavior difference and build its relationship with the policy performance via bf Role Diversity We find that the error bound in MARL can be decomposed into three parts that have a strong relation to the role diversity. The decomposed factors can significantly impact policy optimization on three popular directions.
  arXiv  Detail & Related papers  (2022-06-01T04:58:52Z)
• Celebrating Diversity in Shared Multi-Agent Reinforcement Learning [20.901606233349177]
  Deep multi-agent reinforcement learning has shown the promise to solve complex cooperative tasks. In this paper, we aim to introduce diversity in both optimization and representation of shared multi-agent reinforcement learning. Our method achieves state-of-the-art performance on Google Research Football and super hard StarCraft II micromanagement tasks.
  arXiv  Detail & Related papers  (2021-06-04T00:55:03Z)
• UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers [108.92194081987967]
  We make the first attempt to explore a universal multi-agent reinforcement learning pipeline, designing one single architecture to fit tasks. Unlike previous RNN-based models, we utilize a transformer-based model to generate a flexible policy. The proposed model, named as Universal Policy Decoupling Transformer (UPDeT), further relaxes the action restriction and makes the multi-agent task's decision process more explainable.
  arXiv  Detail & Related papers  (2021-01-20T07:24:24Z)

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.