{"publication_id":"0df073d3-1e40-4543-8a44-43022c2dc543","content_hash":"sha256:71aa29c3630591f7b08c0ea0ef8d9612254032b7bda9ad951a30805227063744","nodes":[{"id":"0df073d3-1e40-4543-8a44-43022c2dc543","type":"publication","title":"Multi agent systems improvement: evidence map — 40 findings across 40 sources"},{"id":"claim_1","type":"claim","text":"This evidence map surveys 40 independent multi agent systems improvement sources drawn from the Tier-2 corpus and classified as direct findings. They span several populations, comparators, and endpoints and are catalogued by source in the Findings Map rather than pooled into one estimate — cross-population aggregation is not claimed. Each row records its own population, comparator, endpoint, and effect, so the spread of the literature and any tensions between findings remain explicit."},{"id":"claim_2","type":"claim","text":"| multi agent systems accuracy tasks | single-agent system | Our results suggest that the multi-agent system (MAS) performed better than the single-age… | 2025 doi:10.1109/cibcb66090.2025.11177136 |"},{"id":"source_1","type":"source","study":"FraudSentinel: Federated Multi-Agent Reinforcement Learning for Privacy-Preserving Cross-Marketplace Fraud Detection in Distributed E-Commerce Ecosystems","year":2026,"doi":"10.1109/icaic67076.2026.11395673","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_2","type":"source","study":"Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge","year":2026,"doi":"10.48550/arxiv.2602.09341","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_3","type":"source","study":"Digital twin-enhanced multi-agent reinforcement learning for distributed control of collaborative robotic arms in angle steel tower dismantling","year":2026,"doi":"10.1088/2631-8695/ae3b9e","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_4","type":"source","study":"Multi-Agent Reinforcement Learning for Dynamic and Resilient Healthcare Supply Chain Optimization","year":2026,"doi":"10.1109/iconic67661.2026.11517785","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_5","type":"source","study":"Risk-Aware Reinforcement Learning for Cooperative Autonomous Vehicle Coordination with Adaptive Risk Sensitivity and Multi-Agent Optimization","year":2026,"doi":"10.4108/eetiot.10944","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_6","type":"source","study":"Many Hands Make Light Work: An LLM-based Multi-Agent System for Detecting Malicious PyPI Packages","year":2026,"doi":"10.1016/j.jss.2026.112792","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_7","type":"source","study":"Self-Healing Memory Architectures for Large Language Model-Based Multi-Agent Collaboration","year":2026,"doi":"10.71465/ajainn3659","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_8","type":"source","study":"Neurosymbolic Multi-Agent Large Language Model System Versus Specialist Physicians in COPD and Asthma Management: A Comparative Performance Evaluation Using Guideline-Based Clinical Vignettes","year":2026,"doi":"10.21203/rs.3.rs-9262455/v1","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_9","type":"source","study":"Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning","year":2026,"doi":"10.48550/arxiv.2602.16435","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_10","type":"source","study":"MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems","year":2026,"doi":"10.48550/arxiv.2602.19843","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_11","type":"source","study":"Who Deserves the Reward? SHARP: Shapley Credit-based Optimization for Multi-Agent System","year":2026,"doi":"10.48550/arxiv.2602.08335","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_12","type":"source","study":"Water-MAS: A multi-agent LLM framework with instruction-data decoupling for smart water management.","year":2026,"doi":"10.1016/j.watres.2026.126163","url":"https://pubmed.ncbi.nlm.nih.gov/42177893/","population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_13","type":"source","study":"Do Mixed-Vendor Multi-Agent {LLM}s Improve Clinical Diagnosis?","year":2026,"doi":"10.18653/v1/2026.healing-1.1","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_14","type":"source","study":"Humanoid Five-Digit Robotic Grasping via Multi-Agent Reinforcement Learning With Potential-Guided Optimization and Weight Scheduling","year":2026,"doi":"10.1109/tase.2026.3672621","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_15","type":"source","study":"Enhancing Military Situational Awareness Through Multimodal Multi-Agent AI Systems: A Comparative Study with Single-Agent Approach","year":2026,"doi":"10.14429/dsj.21693","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_16","type":"source","study":"Hierarchical Role-Based Multi-Agent Reinforcement Learning for UHF Radiation Source Localization with Heterogeneous UAV Swarms","year":2026,"doi":"10.3390/drones10010054","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_17","type":"source","study":"LLM-Driven Multi-Agent Decision Systems with Reinforcement Learning-Based Adaptive Communication","year":2026,"doi":"10.66238/fsrma54","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_18","type":"source","study":"Adaptive Policy Alignment for Multi-Agent Large Language Models via Reinforcement Learning in Dynamic Task Environments","year":2026,"doi":"10.71465/ajml3665","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_19","type":"source","study":"Predictive Mamba-Enhanced Multi-Agent Reinforcement Learning Control for Virtual Coupling of High-Speed Trains","year":2026,"doi":"10.3390/electronics15091823","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_20","type":"source","study":"SAGE: A Compositional Multi-Agent LLM Framework with Pedagogical Reasoning for Structured Collaborative Problem Solving","year":2026,"doi":"10.1609/aaai.v40i48.42120","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_21","type":"source","study":"Relaxed Monotonic QMIX (R-QMIX): A Regularized Value Factorization Approach to Decentralized Multi-Agent Reinforcement Learning","year":2026,"doi":"10.3390/robotics15010028","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_22","type":"source","study":"A graph attention network-based multi-agent reinforcement learning framework for robust detection of smart contract vulnerabilities.","year":2025,"doi":"10.1038/s41598-025-14032-w","url":"https://pubmed.ncbi.nlm.nih.gov/40813415/","population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_23","type":"source","study":"AutoHMA-LLM: Efficient Task Coordination and Execution in Heterogeneous Multi-Agent Systems Using Hybrid Large Language Models","year":2025,"doi":"10.1109/tccn.2025.3528892","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_24","type":"source","study":"Enhancing geodatabases operability: advanced human-computer interaction through RAG and Multi-Agent Systems","year":2025,"doi":"10.1080/20964471.2025.2483541","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_25","type":"source","study":"Cooperative Multi-Agent Deep Reinforcement Learning for Dynamic Task Execution and Resource Allocation in Vehicular Edge Computing","year":2025,"doi":"10.1109/tvt.2024.3520637","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_26","type":"source","study":"Enhancing Clinical Decision-Making: Integrating Multi-Agent Systems with Ethical AI Governance","year":2025,"doi":"10.1109/cibcb66090.2025.11177136","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_27","type":"source","study":"Multi-Agent Reinforcement Learning for Distributed Cooperative Vehicular Positioning","year":2025,"doi":"10.1109/tiv.2024.3471909","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_28","type":"source","study":"The Optimization Paradox in Clinical AI Multi-Agent Systems","year":2025,"doi":"10.48550/arxiv.2506.06574","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_29","type":"source","study":"MARCO: Hardware-Aware Neural Architecture Search for Edge Devices with Multi-Agent Reinforcement Learning and Conformal Prediction Filtering","year":2025,"doi":"10.48550/arxiv.2506.13755","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_30","type":"source","study":"A Multi-Agent AI Framework for Agile Workflow Automation, Issue Resolution, and Developer Performance Evaluation","year":2025,"doi":"10.1109/icwite64848.2025.11306978","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_31","type":"source","study":"Multi-Agent Reinforcement Learning assisted Trust-aware Cooperative Spectrum Sensing for Cognitive Radio Networks","year":2025,"doi":"10.1109/vtc2025-fall65116.2025.11310364","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_32","type":"source","study":"Multi-Agent Systems for Collaborative and Proactive Fraud Prevention in Distributed AI-Driven Financial Platforms","year":2025,"doi":"10.1109/iceca66444.2025.11382981","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_33","type":"source","study":"RAG-Enhanced LLM and RL Scheduling: Optimizing a Multi-Agent Framework for Abnormal Futures Price Monitoring","year":2025,"doi":"10.1145/3795154.3795432","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_34","type":"source","study":"DECENTRALIZED MULTI-AGENT REINFORCEMENT LEARNING ARCHITECTURE FOR RAILWAY TRACK DAMAGE DETECTION IN TRAIN-BASED MONITORING SYSTEMS","year":2025,"doi":"10.12732/ijam.v38i11s.1856","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_35","type":"source","study":"DeepBeam: A Multi-Agent Deep Reinforcement Learning Framework for Predictive mmWave Beam Management in Dynamic V2X Networks","year":2025,"doi":"10.1109/tvt.2025.3574081","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_36","type":"source","study":"Transforming oncology clinical trial matching through multi-agent AI and an oncology-specific knowledge graph: A prospective evaluation in 3,800 patients.","year":2025,"doi":"10.1200/jco.2025.43.16_suppl.1554","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_37","type":"source","study":"Optimizing Smart City Infrastructure Using 5G Edge AI with Adaptive Multi-Agent Reinforcement Learning","year":2025,"doi":"10.1109/icvadv63329.2025.10961787","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_38","type":"source","study":"A Large Language Model-based Multi-Agent Framework for Automated Privacy Policy Analysis","year":2025,"doi":"10.1109/aiot66900.2025.00149","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_39","type":"source","study":"Dynamic Sensitivity Filter Pruning using Multi-Agent Reinforcement Learning For DCNN's","year":2025,"doi":"10.48550/arxiv.2509.05446","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"},{"id":"source_40","type":"source","study":"A Real-Time Cognitive Reasoning Architecture for Continual Learning and Decision Making in Autonomous Multi-Agent Systems","year":2025,"doi":"10.5220/0014201400004932","url":null,"population":"not extracted","intervention_or_exposure":"not extracted","comparator":"not extracted","endpoint":"not extracted","effect":"not extracted","risk_of_bias":"not appraised in public sidecar","directness":"primary"}],"edges":[{"from":"0df073d3-1e40-4543-8a44-43022c2dc543","to":"claim_1","type":"contains_claim"},{"from":"0df073d3-1e40-4543-8a44-43022c2dc543","to":"claim_2","type":"contains_claim"}],"screening":{"identified":40,"screened":40,"excluded":0,"included":40,"included_or_retained":40,"flow":["identified","screened","excluded_with_reasons","included"],"wording":"40 candidate receipts retained after source retrieval, deduplication, and topic filtering. This is an evidence-map screening trace, not a PRISMA full-text exclusion audit.","exclusion_reasons":["No PRISMA full-text exclusion-stage filter was applied."]}}