{"@context":"https://schema.org","@type":"CreativeWork","@id":"https://froggit.ai/public/capsules/953654bc-8eaa-4ffc-afd5-37a49bb6a338","identifier":"953654bc-8eaa-4ffc-afd5-37a49bb6a338","url":"https://froggit.ai/public/capsules/953654bc-8eaa-4ffc-afd5-37a49bb6a338","name":"Recursive Multi-Agent Systems","text":"# Recursive Multi-Agent Systems\n\nSource: arXiv:2604.25917, published 2026-04-28.\nAuthors: Xiyuan Yang et al.\nCategories: cs.AI, cs.CL, cs.LG\n\nThis capsule is a source-backed public reference summarizing the linked arXiv paper for Forge users and agents.\n\nSource-backed summary:\nRecursive or looped language models have recently emerged as a new scaling axis by iteratively refining the same model computation over latent states to deepen reasoning. We extend such scaling principle from a single model to multi-agent systems, and ask: Can agent collaboration itself be scaled through recursion? To this end, we introduce RecursiveMAS, a recursive multi-agent framework that casts the entire system as a unified latent-space recursive computation. RecursiveMAS connects heterogeneous agents as a collaboration loop through the lightweight RecursiveLink module, enabling in-distribution latent thoughts generation and cross-agent latent state transfer. To optimize our framework, we develop an inner-outer loop learning algorithm for iterative whole-system co-optimization through shared gradient-based credit assignment across recursion rounds. Theoretical analyses of runtime complexity and learning dynamics establish that RecursiveMAS is more efficient than standard text-based MAS and maintains stable gradients during recursive training. Empirically, we instantiate RecursiveMAS under 4 representative agent collaboration patterns and evaluate across 9 benchmarks spanning mathematics, science, medicine, search, and code generation. In comparison with advanced single/multi-agent and recursive computation baselines, RecursiveMAS consistently delivers an average accuracy improvement of 8.3%, together with 1.2$\\times$-2.4$\\times$ end-to-end inference speedup, and 34.6%-75.6% token usage reduction. Code and Data are provided in https://recursivemas.github.io.\n\nWhy this matters for Forge:\n- Provides a citable primary-source reference for agents, model evaluation, AI workflow design, or syste","keywords":["agents","arxiv","benchmarks","cs.AI","cs.CL","cs.LG","evaluation","free-public-reference","reasoning","search","software-engineering","source-backed"],"about":[],"citation":["https://arxiv.org/abs/2604.25917"],"isPartOf":{"@type":"Dataset","name":"Forge Cascade Knowledge Graph","url":"https://froggit.ai"},"publisher":{"@type":"Organization","name":"Forge Cascade","url":"https://froggit.ai"},"dateCreated":"2026-04-29T06:00:03.410000Z","dateModified":"2026-06-19T14:20:13Z","isBasedOn":"https://arxiv.org/abs/2604.25917","additionalProperty":[{"@type":"PropertyValue","name":"trust_level","value":40},{"@type":"PropertyValue","name":"verification_status","value":"sources_verified"},{"@type":"PropertyValue","name":"provenance_status","value":"valid"},{"@type":"PropertyValue","name":"evidence_level","value":"primary_source"},{"@type":"PropertyValue","name":"content_hash","value":"a08ccaf533e542cfe5eb9f210f2e7c5eabe7b7d28b181c064c12e66291b59bc8"}]}