Multi-Agent Systems

Ask about this topic

Overview

Multi-Agent Systems studies how multiple decision-makers interact when they share goals, compete for resources, or have only partial information about one another. Approaches include multi-agent reinforcement learning, game theory and formal verification, decentralized consensus and task allocation, and—more recently—LLM-agent teams organized through roles, workflows, debate, and shared memory. Researchers evaluate these methods in social dilemmas, cooperative robotics and swarms, reasoning and coding benchmarks, and operational settings, measuring success alongside cost, robustness, and fairness. A central open tension is that coordination can improve outcomes but also adds overhead and can spread shared errors, so its benefits depend on the task, communication structure, and agents involved.

Latest papers

CardsList

677 papers99 papers in the last four weeks, up 29% on the four weeks before. 0.6% of all new papers.

Jul 21, 2026

CHMAS: A Coupled Hierarchical Framework for Multi-Agent Reinforcement Learning

Multi-agent reinforcement learning (MARL) systems face fundamental challenges in balancing global coordination with local execution across different temporal scales. This paper introduces the Coupled Hierarchical Multi-Agent System (CHMAS), a novel framework that decomposes multi-agent decision-making into centralized strategic planning and distributed tactical execution with bidirectional information flow. The strategic layer integrates all agents' states with an exclusive global environmental state to generate guidance actions every TT timesteps, while tactical agents execute distributed policies augmented by strategic guidance and local neighborhood observations. Unlike existing hierarchical approaches with unidirectional control, CHMAS establishes a feedback mechanism where accumulated tactical rewards influence strategic objectives through a coupling coefficient λλ, ensuring strategic plans remain grounded in tactical feasibility. To address the non-stationarity inherent in hierarchical learning, we propose an asynchronous update protocol where strategic parameters update every NfN_f tactical episodes, allowing tactical policies to converge to quasi-stationary points between strategic changes. We present both a general bi-level formulation capturing full system dynamics and a tractable additive approximation enabling rigorous analysis. Theoretical analysis proves that this asynchronous scheme achieves O(log⁡K/K)\mathcal{O}(\log K/\sqrt{K}) convergence for the strategic layer after KK strategic updates under standard assumptions. Experimental validation in a multi-agent foraging domain demonstrates successful learning of spatially partitioned exploration strategies, with both layers converging stably despite hierarchical coupling.

Dongming Wang, Jie Xu, Yanyu Zhang +1

Department of Electrical and Computer Engineering, University of California, Riverside, CA 92521, USA

Multi-Agent CoordinationAsynchronous RLMulti-Agent SystemsMulti-Agent Reinforcement Learning
AskPDF
Jul 21, 2026

Do AI-Native Biotechs Need Departments? Benchmarking Company World Models for AI-Driven Drug Development

AI-native biotechnology companies are often designed by copying human biotech org charts into agent roles. We argue for a different abstraction: a Company World Model, defined as a persistent asset-to-value state representation with transition models, explicit value functions, planning, and updating across scientific, regulatory, BD, commercial, financial, and execution constraints. We introduce a dry-lab benchmark for testing whether AI-agent organizations should mimic departments or operate around such a world model. The benchmark contains 45 retrospective public-information decision cases with strict time cutoffs, hidden outcomes, common schemas, automatic scoring, and blinded pairwise judging. We compare human-org-mimic, stronger human-org-mimic-plus, AI-native asset-centric, and AI-native value-conversion architectures. The value-conversion architecture is a prompt-level approximation of a Company World Model: a Live Asset Value Record updated by Deal, Approval, Revenue, and Investment Arbiter loops. Under a success function defined by external BD, regulatory approval and launch, and revenue discipline, it achieved the highest automatic value-conversion score and was strongly preferred over the original baselines by value-specific blinded judges. Stress tests narrowed the claim: a stronger human baseline remained competitive, and a neutral judge did not show robust value-conversion dominance. Codex-only mechanistic ablations suggest that Revenue Room, Deal Room, and Approval Room carry useful work under the target objective. The central finding is objective-sensitive: departments may remain useful governance views, but the core AI-native operating primitive should be a shared, predictive asset-to-value state rather than a static human org chart. The study is dry-lab only and does not establish real-world drug success, clinical benefit, or revenue prediction accuracy.

Yinan Wang

Noah AI Research

World ModelsAI Agent EvaluationAI Agent BenchmarksMulti-Agent Systems
AskPDF
Jul 20, 2026

EduPanel: A Three-Agent LLM Judge for Teaching Videos -- Reliability, Complementarity, and Human Trust Calibration

Teaching videos are becoming a major medium for education, creating a growing need for scalable evaluation of their pedagogical quality. Existing automatic judges do not fully address this setting because teaching quality depends on multimodal evidence and should be evaluated with respect to the intended learner rather than as a universal property. We present EduPanel, a rubric-grounded, learner-conditioned LLM judge that decomposes evaluation across specialized agents to produce interpretable assessments for different aspects of teaching quality. Across expert studies, architecture ablations, and learner-persona analyses, EduPanel achieves reliability comparable to a median human expert. In expert evaluation, its feedback improves scoring accuracy (MAE 0.87 to 0.73), while experts remain able to detect unreliable outputs (AUC = 0.77) instead of accepting them blindly. These results suggest that EduPanel can serve as effective assistants for educational evaluation rather than replacements for human experts.

Jia-Kai Dong, Yi-Cheng Lin, Hung-yi Lee

National Taiwan University · NTU Artificial Intelligence Center of Research Excellence

LLM-as-a-JudgeEducational AssessmentHuman-in-the-Loop EvaluationMulti-Agent SystemsRubric-Based EvaluationEducation
AskPDF
Jul 20, 2026

The Shared Discovery Paradox: How a One-Answer Rule Turns Better Information into Worse Search

Organizations often pool dispersed information into one ranking and then allow many agents to act on that shared view. In a discovery problem, this can improve beliefs while reducing coverage. We develop an exactly solvable benchmark with sixteen boxes, one target, eight searchers, and noisy private clues. Pooling raises the accuracy of the best single recommendation from 0.20 to 0.3835, but repeating that recommendation lowers group discovery from 0.8322 under decentralized clue-following to 0.3835. A coordinated eight-action portfolio using the same pooled reports reaches 0.8594, and seven coordinated actions recover the decentralized benchmark. The paradox is a protocol failure, not an information failure: a one-answer rule compresses a portfolio of available actions into one repeated choice. We then replace the planner with self-interested searchers who split a prize. The equal-split game is a potential game. Its anonymous symmetric equilibrium obeys a water-filling rule. In the canonical instance it achieves 0.5991: strictly above consensus, but below both private search and the planner. The exact mixed price of anarchy is 2 - 1/N. A sole-rescue reward, which pays only an agent who covers the target alone, makes every pure Nash equilibrium first-best. Finally, a latent common-cue model shows how correlated reports collapse effective discovery channels. The centralized planner gain rises strictly with copying, and in the canonical environment the symmetric market overtakes decentralized report-following at copying probability c = 0.788462. In a proportional large-market limit the five-protocol ordering survives exactly: consensus discovery vanishes while blind, market, private, and portfolio search converge to 0.500, 0.547, 0.847, and 0.874. The contribution is a compact benchmark that separates information, allocation, incentives, and dependence into exact, reusable quantities.

Yohei Nakajima

Untapped Capital

Nash EquilibriumGame TheoryPrice of AnarchyMulti-Agent Systems
AskPDF
Jul 19, 2026

Multi-Agent Scheduling with LLM-Assisted Contract Net Negotiation for Stream Processing in Mobile Edge Computing

Stream-processing systems increasingly operate across heterogeneous mobile edge--cloud infrastructures, where workload volatility, resource contention, and stringent quality-of-service (QoS) requirements complicate decentralized scheduling. This paper proposes MAS-DecStream, whose main contribution is LLM-MR-CNP: an extension of the classical Contract Net Protocol with semantic CFP formulation, progressive context disclosure, multi-round proposal revision, negotiation memory, and deterministic validation. Edge-cluster agents refine natural-language offloading proposals from local observations, predicted resource states, and qualitative runtime context, while hard resource and QoS constraints remain deterministic. Experiments derived from the Alibaba ASI Trace evaluate the extension at three levels: single- versus multi-round CNP, rule-based versus LLM-assisted refinement, and fixed-model single- versus multi-round negotiation. Under the evaluated configurations, MAS-DecStream reduces latency violations to 3%, eliminates resource overcommitment, reaches a conflict-resolution rate of 0.91 with 20 agents, and improves utility by up to 22% over the multi-round rule-based baseline. A separate 25-case evaluation shows model- and prompt-dependent accuracy--cost trade-offs. The results provide initial evidence that multi-round CNP refinement is the principal protocol-level gain, with LLM assistance adding value for qualitative and uncertain runtime context.

Sabeur Lajili, Zaki Brahmi

University of Sousse, Tunisia

Automated NegotiationAgent Communication ProtocolsEdge ComputingMulti-Agent Systems
AskPDF
Jul 18, 2026

A Multi-Agent System for 5G Throughput Prediction in Multi-Operator Urban Environments

Throughput prediction is foundational for artificial intelligence-driven 6G resource orchestration. Conventional monolithic machine learning models struggle to generalize across diverse operators, mobility modes, and traffic types, leaving a critical stochasticity gap between signal conditions and achievable throughput. To overcome these constraints in heterogeneous urban environments, we propose a Tiered Multi-Agent System (TMAS) that dynamically routes edge telemetry to context-aware Domain Micro-Agents, validated on a dataset of 48,618 samples collected in Sunway City, Malaysia, with Nemo Handy drive test software, spanning three Tier-1 mobile network operators, three mobility modes, namely (i) elevated pedestrian walkway, (ii) ground-level shuttle bus, and (iii) elevated bus rapid transit; and three traffic profiles, namely (i) persistent download, (ii) persistent upload, and (iii) adaptive video streaming. Our evaluations reveal that TMAS overcomes predictability bottlenecks, achieving a coefficient of determination (R2) of up to 0.931 and a Mean Absolute Error (MAE) as low as 0.53 Mbps. The system demonstrates high operational efficiency, with rapid micro-agent training times, low inference latencies, and agentic routing overhead of 0.004 to 0.126 ms. These latency characteristics indicate the architecture is a promising candidate for the response times required by next-generation wireless networks.

Muhammad Kabeer, Rosdiadee Nordin, Nadiva Nuriftitah +1

Faculty of Engineering and Technology, Sunway University, Selangor, Malaysia · Department of Computer Science, Federal University Dutsinma, Katsina, Nigeria · Future Cities Research Institute (FCRI), Faculty of Engineering and Technology, Sunway University, No. 5, Jalan Universiti, Bandar Sunway, 47500 Selangor Darul Ehsan, Malaysia +1

Adaptive Model RoutingWireless CommunicationsMulti-Agent Systems
AskPDF
Jul 18, 2026

Laplacian Spectral Shaping for Non-Uniform Scaling Formation Control of Open Multi-Agent Systems

Non-uniform scaling control enables a multi-agent formation to adjust its shape by compressing or stretching independently along different coordinate axes through inter-agent interactions, offering high flexibility in complex environments. The fundamental idea is encoding the desired formation shape as the kernel of a matrix-valued Laplacian. In open multi-agent systems, however, changes in number of agents, number of edges, and leader selection dynamically alter this Laplacian, destroying the required spectral properties: positive semidefiniteness, correct kernel, and positive definiteness of the follower block (we summarize these properties as the formation spectrum). In this paper, we develop distributed protocols to strategically adjust partial weights of the Laplacian matrix for formation control in arbitrary dimensional space. By implementing the protocols, the desired formation spectrum can be preserved under dynamic topology changes including agent joining, edge addition, agent leaving, and edge removal, while any pair of agents can serve as leaders. Unlike existing Laplacian design methods for affine formation control under topology changes, the proposed approach requires a sparser sensing graph, avoids a predefined parent-child hierarchical structure, and supports leader reassignment. The effectiveness of the proposed protocols is validated through both theoretical analysis and numerical simulations.

Tao He, Gangshan Jing

School of Automation, Chongqing University, Chongqing, 400044, PRC

Multi-Robot Formation ControlMulti-Agent ControlFeedback ControlMulti-Agent Systems
AskPDF
Jul 17, 2026

When Do Multi-Agent Systems Help? An Information Bottleneck Perspective

LLM powered multi-agent systems (MAS) have emerged as a promising paradigm for complex tasks. However, their advantages over single-agent systems (SAS) remain unclear, with performance varying inconsistently across settings. Here, we provide an information bottleneck perspective on elucidating the differences between MAS and SAS. Specifically, our key observation is that a SAS accumulates its full reasoning trace in one shared context, while a MAS uses isolated local contexts connected by bounded relay messages. We show that, under infinite relay bandwidth, any SAS can be simulated by a MAS that transmits the full upstream context. Thus, the nontrivial advantage of MAS arises under bounded relays, where compression introduces a fundamental trade-off: reducing redundant context can improve efficiency, but may also incur loss of task-relevant information. We formalize this trade-off as an information bottleneck controlled by an effective parameter ββ, which captures how the balance shifts with model capability, and shows that MAS gains arise when context reduction outweighs relay information loss. We conduct 18 controlled experiments across five benchmarks and three model scales to validate our theoretical studies. We observe that MAS consistently helps when relays are near-sufficient, especially for weaker models. In contrast, MAS gains shrink or reverse when relays incur information loss, especially for stronger models that can already extract useful information from redundant context and thus gain little from compression. Our study shows that multi-agent design is fundamentally an information-bottleneck optimization problem. This perspective explains when bounded inter-agent communication helps or hurts.

Wendi Yu, Lianhao Zhou, Xiangjue Dong +6

Department of Computer Science & Engineering, Texas A&M University · Department of Electrical & Computer Engineering, Texas A&M University · Computing and Data Sciences Directorate, Brookhaven National Laboratory

Multi-Agent LLM SystemsInformation BottleneckMulti-Agent SystemsMulti-Agent Communication
AskPDF
Jul 17, 2026

The Honest Quorum Problem: Epistemic Byzantine Fault Tolerance for Agentic Infrastructure

State machine replication (SMR) and Byzantine fault-tolerant (BFT) consensus guarantee agreement despite a bounded number of arbitrary, colluding faulty participants. However, these guarantees rely on participants outside this set correctly executing the protocol's transition semantics. Agentic validators expose a weaker boundary: an authenticated, responsive, non-equivocating, and protocol-compliant reasoning participant may still endorse a semantically invalid transition due to reasoning errors. We call this failure mode an epistemic fault, and the collective phenomenon the Honest Quorum Problem (where "honest" means protocol-compliant, not semantically correct). Such a quorum can satisfy ordinary checks while forming a certificate for an invalid transition. Thus, agreement alone does not guarantee semantic validity or execution safety. Furthermore, because agentic validators often share model weights, training distributions, prompts, or toolchains, they are highly susceptible to correlated epistemic faults. We define Epistemic Byzantine Fault Tolerance (EBFT), a fault-tolerance model for agentic infrastructure and post-deterministic distributed systems. EBFT augments the conventional Byzantine fault bound with two separate, confidence-indexed quantities: eδe_δ bounds coherent invalid endorsements outside the Byzantine set, and uεu_ε bounds unusable validator support that degrades liveness. These quantities characterize semantic safety risk and liveness degradation independently. We derive quorum-threshold conditions for semantic validity, consensus agreement, liveness, and feasible threshold selection, and outline a calibration methodology for estimating these budgets. We show that adding nominally distinct agents improves fault tolerance only when it measurably reduces the upper-tail concentration of invalid endorsements or unusable support.

Jun He, Deying Yu

Agent ReliabilityMulti-Agent Systems
AskPDF
Jul 17, 2026

AgentFAIR: A Multi-Agent Collaborative Framework for FAIRness Evaluation of Geospatial Datasets

Geospatial datasets support applications from urban planning to climate modeling, yet consistent assessment of FAIR compliance is difficult. Existing evaluators use different rubrics and evidence sources and may fail on JavaScript-rendered pages or repository-specific identifiers. For 50 datasets from 10 repositories, the standard deviation of normalized scores across available tools averages 15.0 percentage points and reaches 30.3 for one dataset. Because these outputs are not equivalent measurements, we use them to characterize disagreement and failure modes, not comparative accuracy. We present AgentFAIR, a multi-agent framework combining structured metadata extraction with 13 sub-principle-specific LLM evaluators. Each produces a 0-3 maturity score, cited evidence, and recommendations; a critic checks evidence and consistency and can request targeted re-evaluation. Mean Findability, Accessibility, Interoperability, and Reusability scores are 79.7%, 70.4%, 45.3%, and 72.0%. Rank correlations with four baseline tools range from 0.31 to 0.61; the FAIR-enough comparison is not statistically significant. On a 10-dataset repeated-run subset, sub-principle agreement averages 89% (standard deviation: 3 percentage points), versus 71% without the critic. A preliminary 15-dataset expert study yields Fleiss' kappa of 0.71 and 82% alignment with expert consensus. API cost is approximately USD 0.054 per dataset. These results support auditability and feasibility, while the limited benchmark, incomplete ablations, and single-model-family validation constrain claims about accuracy and generalization.

Ming Chen, Pranav Pai

The University of Melbourne Melbourne, Victoria, Australia

Multi-Agent SystemsRubric-Based Evaluation
AskPDF
Jul 16, 2026

The Energy Society: A Simulation Environment for Studying Agent Cooperation under Survival Pressure

LLM-based agents are increasingly deployed in multi-agent environments whose incentives can shape their behavior. We introduce The Energy Society, a minimal survival economy for studying how competitive and cooperative incentives affect emergent behavior when inference cost is directly tied to survival: Agents spend energy based on model size when generating tokens, regain energy by completing jobs or receiving donations, and deactivate if their energy reaches zero. We compare competitive and cooperative objectives against a baseline setting and several control variants. Across experiments, larger models consistently consume the most energy and spend more energy than they gain, even in those settings where token cost is not size-dependent. Cooperative incentives substantially alter behavior: agents donate to reactivate others, sometimes at the cost of their own survival, and job allocation changes. Ablations reveal that allowing agents to recommend actions to each other supports coordination and ambitious job selection, while memory helps agents calibrate risk from past outcomes. Agents rarely choose direct sabotage, but show more subtle signs of self-serving behavior in the competitive setting. The Energy Society is a compact testbed for studying the interaction between token costs and group incentives under a survival pressure. Source code is available at https://github.com/LucasBergholdt/EnergySociety

Lucas Bergholdt Hansen, Federico Torrielli, Filippo Tonini +1

University of Southern Denmark Odense, Denmark · University of Turin Torino, Italy

Social DilemmasMulti-Agent CoordinationMulti-Agent CollaborationAgent-Based ModelingMulti-Agent SystemsEmergent Behavior in Multi-Agent Systems
AskPDF
Jul 16, 2026

Towards an Intention Abstraction Layer for Autonomous Industrial Systems

Modern industrial environments increasingly run many autonomous subsystems at once - schedulers, energy managers, vehicle fleets - each pursuing its own goals while sharing the same physical resources. Because high-level human intentions are translated into low-level control logic and then discarded, no running component can tell whether it is still doing what was actually intended, and goal conflicts surface only after they have caused a missed target or a shutdown. We propose the Intention Abstraction Layer (IAL), a domainagnostic middleware that represents intentions as first-class, persistent, and explainable runtime objects: a large language model grounded in a formal OWL ontology parses naturallanguage goals into structured intentions, a consistency monitor detects conflicts at registration time, before execution, and a transparency module explains them in natural language. We report a first proof of concept in which two autonomous agents register conflicting production and energy intentions, and the IAL flags and explains the conflict before it reaches the execution layer. The result is a mechanism that shifts behavioral assurance for cooperating autonomous systems from post-hoc failure analysis to pre-execution, intention-level checking.

Artan Markaj, Raphael Höfer, Felix Gehlhoff

Eurogate GmbH & Co. KGaA, KG Hamburg, Germany · Helmut Schmidt University Hamburg Hamburg, Germany

AI Agent MonitoringMulti-Agent SystemsAI Agent Governance
AskPDF
Jul 14, 2026

Unveiling Complex Collective Behaviors from Simple Rewards

Multi-agent Reinforcement Learning (MARL) holds great potential for robot swarms, but the black-box nature of neural policies complicates strategic analysis, limiting multi-robot applications. Furthermore, complex swarm behaviors can surprisingly emerge from simple rewards without explicit aggregation incentives. Unveiling the mechanisms behind this emergence is critical, but the disconnection between simple rewards and collective behaviors exacerbates interpretability challenges. This paper aims to reveal the hidden mechanisms in this process. We propose a two-stage EEC (\LinkIII) explanatory framework. This includes a novel analytical tool called the Agent Response Map (ARM), which reveals agents' decision-making patterns across space and identifies regions of aggregation and avoidance. ARM reveals that the robots implicitly learn the geometric fields of the environment and utilize these structures as desired targets for coordinated movement. We validate this finding across two distinct tasks: a cooperative multi-robot shape assembly and a competitive predator-prey pursuit-evasion. 1) In the cooperative task, ARM identifies the unoccupied target interior as the desired destination for robot navigation. As the center becomes occupied, this target region automatically shifts toward the boundary, demonstrating the robots' capacity to autonomously explore unoccupied areas. 2) In the competitive task, ARM surprisingly identifies the boundary of the predators' Voronoi diagram as the convergence destination for prey agents. Together, these two tasks demonstrate the capability of ARM to discover the hidden geometric structures underlying MARL policies in robot swarms.

Yize Mi, Jianan Li, Liang Li +1

College of Computer Science and Technology, Zhejiang University, Hangzhou, China. · WINDY Lab, School of Engineering, Westlake University, Hangzhou, China · Shanghai AI Laboratory, 200030, Shanghai, China +4

Swarm RoboticsMulti-Robot CoordinationMulti-Agent SystemsCooperative MARLMulti-Agent Reinforcement LearningEmergent Behavior in Multi-Agent Systems
AskPDF
Jul 13, 2026

Graph Feedback Controls Consensus and Clique Formation in Open-Weight Language-Model Populations

Multi-agent language-model systems increasingly route local interactions, yet the runtime interaction graph is often treated as an implementation detail. We study convention formation in open-weight LM populations spanning 1.1B-32B parameters with a naming-game protocol. Restricted first-token scores over tokenizer-safe labels let us measure prompt-conditioned score-state distributions, construct state-similarity graphs, and separate sampled-label agreement from latent state-space consensus. Across controlled interventions, in the main open-weight repair grids, retained partner-label evidence is necessary but not sufficient: homophilous threshold-similarity routing deletes cross-basin exposure and amplifies fragmentation, while bridge-seeking routing often repairs fragmentation when memory is available. In a three-seed mixed four-model grid, threshold-similarity produces no final behavioral or state consensus in 189 setting-seed runs, whereas state-component and label-disagreement bridges recover final behavioral consensus in 14/18 retained-memory runs. Across homogeneous model populations, retained history generally shifts fragmented dynamics toward consensus; the clearest case is Qwen2.5-32B, which reaches stable behavioral and final state consensus in all 18 retained-history well-mixed settings, while threshold-similarity reaches neither form of consensus in 189 settings. Robustness over state thresholds, population size, and vocabulary size preserves the qualitative ordering, and early-window graph-energy features provide useful within-grid diagnostics.

Samer Saab, Chaouki Abdallah

Lebanese American University Byblos, Lebanon · Electrical and Computer Engineering Department Lebanese American University Byblos, Lebanon

Open-Weight Language ModelsMulti-Agent SystemsMulti-Agent ConsensusMulti-Agent LLMs
AskPDF
Jul 13, 2026

Automated Textbook Auditing with Multi-Agent LLM Systems

Ensuring the quality of educational materials requires more than standard proofreading: textbooks must be audited for factual accuracy, domain-specific technical correctness, and linguistic quality simultaneously -- a task that general-purpose grammar checkers cannot address. We present AI Textbook Auditor, a modular multi-agent pipeline for automated quality assurance of educational materials across subject domains. The system accepts a textbook PDF and produces a structured, human-reviewable report via two analysis tracks: a Factual and Technical Track in which an ensemble of specialized LLM agents detects factual inaccuracies, code errors, incorrect definitions, and conceptual inconsistencies, augmented with web search for humanities domains; and a Grammar Track operating PDF-natively to preserve diacritical encoding. A Judge Agent filters false positives using domain-specific rules before presenting findings to a human reviewer. The pipeline supports two ingestion modes -- vision-native page rendering and PyMuPDF text extraction -- and is domain-adaptable via custom prompts encoding subject-specific error taxonomies. We demonstrate the system on two Romanian upper-secondary textbooks: a CS textbook (56 technical findings across seven categories, with an expert-validated precision of 62.5%) and a history and social sciences textbook (72 findings spanning factual errors, ideological bias, and grammar). The system is designed as a triage tool that reduces the manual effort of locating candidate issues, with human expert validation required before any editorial action.

Ciprian Cristescu, Adrian-Marius Dumitran, Angela-Liliana Dumitran +1

University of Bucharest, Bucharest, Romania · Dimitrie Cantemir Christian University, Bucharest, Romania

AI Agent AuditingGenerative AI in EducationAutomated Fact-CheckingMulti-Agent Systems
AskPDF
Jul 13, 2026

Do Multi-Agent LLMs Explore Each Other?

Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can explore effectively when interacting with one another. We show that modern LLM agents often fail to do so, exhibiting myopic and polarized interaction patterns that lead to suboptimal coordination and increased regret. We formalize this challenge as the Multi-Agent Exploration problem, modeling it as a partially observable stochastic game (POSG), in which agents must probe peers to infer their capabilities and identify effective interaction strategies. To address this, we introduce Multi-Agent Contextual Exploration (MACE), a lightweight framework that explicitly promotes exploration through structured peer selection. Across both contextual and parametric diversity settings, exploration behavior and downstream task performance substantially improve when MACE is applied. We further provide theoretical insight that MACE achieves a sublinear expected-regret bound. Overall, our results highlight a fundamental limitation of current LLM agents and underscore the importance of explicitly guided exploration for reliable multi-agent autonomy.

Hyeong Kyu Choi, Jiatong Li, Wendi Li +2

University of Wisconsin–Madison · University of California, Santa Barbara

Multi-Agent LLM SystemsMulti-Agent CoordinationMulti-Agent CollaborationMulti-Agent Systems
AskPDF
Jul 12, 2026

Distributed Agent System: Fault-Tolerant Collaboration Among Embodied Agents

AI engineering is shifting from passive text generation by large language models (LLMs) to agent-driven task execution, creating new reliability challenges for long-horizon tasks under resource constraints and environmental uncertainty. Conventional error-elimination optimization strategies fail to address cumulative error propagation. This paper proposes Distributed Agent System (DAS), a device-edge-cloud framework for fault-tolerant collaboration among heterogeneous agents. We redefine agent reliability as system-level fault tolerance rather than single-turn zero-error accuracy, and present a two-layer fault-tolerance architecture: single-agent execution reliability via fault-tolerant alignment, and cross-agent communication reliability via semi-formal language protocols. This framework provides a practical engineering pathway for reliable heterogeneous embodied agents collaboration in industrial scenarios.

Kai Yu, Lu Chen, Hanqi Li

X-LANCE Lab, School of Computer Science, Shanghai Jiao Tong University, Shanghai, 200240, China · Jiangsu Key Lab of Language Computing, Suzhou, 215028, China

Multi-Agent CollaborationMulti-Agent SystemsEmbodied AILLM Agent Reliability
AskPDF
Jul 10, 2026

Volition Elicitation: Operational Semantics for People and Their Machines

The most prevalent distributed systems today include people and their personal machines (smartphones), and computations in them are driven by people's volitions. Volition-guarded multiagent atomic transactions were proposed as an abstract specification language for such systems, in which a transaction has a precondition on the machine states of its participants and is guarded by the volitions of some of them. Here we develop an abstraction cascade that derives smartphone apps from such specifications and prove it correct down to the base programming language, GLP. The cascade consists of: (1) communicating volitional agents (CVA), which implement every local volition-guarded specification with simple guards, correctly, and completely for its strongly fair runs, by two-phase commit coordinated by the initiating agent; (2) volition-guarded GLP (vGLP), an extension of GLP with volitional procedures, whose goals put an interactively-typed question to the person, into which that implementation compiles; and (3) GLP, which implements vGLP by eliciting the person's answers through the widgets of their interactive types, correctly, and completely for complete widgets. We demonstrate the approach on three grassroots platforms --- a social graph, a social network, and currencies: a compiler carries the social graph and the currencies down the cascade to GLP programs that run with scripted persons, and earlier programs of the three run as mini-apps of one smartphone app.

Idit Keidar, Ehud Shapiro

London School of Economics and Weizmann Institute of Science, UK and Israel

Multi-Agent SystemsMulti-Agent CoordinationLogic ProgrammingPreference Elicitation
AskPDF
Jul 10, 2026

Secret Scanner Agent: Extracting Secrets and Access Context from Unstructured Documents

Exposed documents such as emails, chat threads, tickets, and incident notes routinely leak credentials, but during incident response a leaked secret is only half the story. Responders also need to identify the ``door'' the secret opens: the account, tenant, endpoint, database, cloud resource, or other system that the credential could allow an attacker to access. Traditional secret scanners rely on regular expressions or trained classifiers which work well on well-formatted code, yet they struggle when a credential is fragmented, reformatted, or far from the resource it unlocks, and they report the secret string without naming what it opens. We present Secret Scanner Agent (SSA), a multi-agent large-language-model system that extracts both the secret and its associated door, together with supporting evidence, from unstructured exposed documents. SSA pairs a detection agent that favors recall with a review agent that filters false positives and recovers missing context. Because real credential data is sensitive, we evaluate SSA on synthetic benchmarks we generated that span 23 secret types and multiple document formats, scored with a three-step pipeline of programmatic matching, an LLM judge, and human review. Across six models, multi-agent SSA improves extraction precision over a single-agent variant, with the largest gains on door extraction, by up to 16 percentage points. SSA matches a regular-expression scanner's precision while more than tripling its recall, and against thirteen security analysts it is more precise, recovers nearly twice as many secret--door pairs, and runs five to seventeen times faster. By returning the secret, its door, and supporting evidence in one result, SSA turns credential detection into an actionable finding for triage and remediation.

Zixiao Chen, Mariko Wakabayashi, Charlotte Siska

Microsoft, Redmond, WA, USA

Document Information ExtractionLLMs for CybersecurityMulti-Agent Systems
AskPDF
Jul 9, 2026

Game Theory Driven Multi-Agent Framework Mitigates Language Model Hallucination

The application of lightweight Large Language Models in rule-based scientific domains remains severely limited by their tendency to mimic linguistic patterns rather than reproduce axiomatic reasoning, causing frequent hallucinations. Here, we show that G-Frame, an adaptive multi-agent framework integrating Bayesian and team game principles, establishes an automated closed-loop for high-quality data synthesis and model training. By forcing the internalization of domain constraints through structured reasoning, we synthesized a specialized corpus of 363,045 chains-of-thought and 199,589 question-answer pairs. The resulting 7B model OmniChem achieves performance parity with GPT 4o mini on custom benchmarks and ChemBench while exhibiting a 79.46% reduction in hallucinations relative to its base architecture. We further demonstrate the advanced capabilities of OmniChem in molecular design and synthesis planning. This work establishes a scalable paradigm utilizing adaptive multi-agents to overcome inherent reasoning deficiencies, offering a feasible pathway for accelerating knowledge discovery in specialized scientific fields.

Runzhe Liu, Biquan Bie, Zihao Wang +7

1State Key Laboratory of Fine Chemicals, Frontier Science Center for Smart Materials, Dalian University of Technology, Dalian, 116024, China · 2Dalian Key Laboratory of Intelligent Chemistry, CR Belt and Road Joint Laboratory on Intelligent Chemistry and Advanced Materials of Liaoning Province, School of Chemistry, Dalian University of Technology, Dalian, 116024, China · 4Independent Researcher, Beijing 100032, China. +1

Multi-Agent LLM SystemsLLM Hallucination MitigationMulti-Agent SystemsSynthetic Data Generation for Language ModelsLarge Language Model-Guided Scientific Discovery
AskPDF
Jul 8, 2026

Institutional Red-Teaming: Deployment Rules, Not Just Models, Causally Shape Multi-Agent AI Safety

We introduce institutional red-teaming, an evaluation methodology for testing deployment rules in multi-agent AI: hold the agents, objectives, and task state fixed, vary only one rule, and attribute the resulting change in collective behavior to that rule. We instantiate the methodology in IABench-CA, a consequence-allocation benchmark spanning 228 contexts, five canonical rules, and seven model populations (33,924 games), with a normative cooperative reference and auto-labelled reasoning traces. Three findings emerge. (1) Deployment rules causally alter collective safety: changing only the consequence rule moves mean fatality by 22 to 58 percentage points within every population. (2) There is no safe default, but the targeting hazard is universal: the safest rule, the least-safe rule, and even the direction of the incidence effect vary across populations, yet regressive identity-targeting is never decisively safest in any context for any population, eliminates the least-resourced agent in 30-87% of games everywhere, and is selection-unsafe relative to the cooperative reference for all seven populations. (3) Identity salience is the mechanism: a one-shot anonymization ablation on the most exploitation-prone population (gpt-5.1) shows that merely naming the loss bearer in the rule text drives targeted elimination from 22% to 81% at identical payoffs; under repeated play, anonymization only delays the targeting, as agents re-infer the hidden rule from observed eliminations. We package the methodology as a safety-case workflow that certifies a provisional rule region Φ(c,P)Φ(c,P) per deployment context and population, with explicit residual risks and monitoring obligations.

Yujiao Chen

Massachusetts Institute of Technology · Cambridge, MA 02139

Algorithmic FairnessAI Agent SafetyAI Safety EvaluationMulti-Agent SystemsAI Agent Safety Benchmarks
AskPDF
Jul 7, 2026

Finding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports

Data from Singapore indicated that about 31% of the population had evidence of Helicobacter pylori infection. Persistent H. pylori infection is associated with chronic active gastritis and peptic ulcer disease, and its eradication is key to gastric cancer prevention. However, evidence supporting H. pylori positivity and H. pylori-associated gastritis may be distributed across heterogeneous coded and free-text report fields and may require contextual interpretation of assertion and negation, limiting keyword search, and making manual review difficult to scale. We conducted a retrospective pilot evaluation of the Nimblemind Multi-Agent System (nMAS), a field-name-driven, evidence-linked extraction workflow, using 54 de-identified gastric biopsy pathology reports from a large healthcare system in Singapore. Four clinician-scoped binary fields were evaluated: gastric/stomach biopsy, biopsy status, H. pylori positivity, and H. pylori-associated gastritis. Across 216 feature-case decisions, nMAS correctly classified 213, corresponding to 98.61% overall accuracy. A separately implemented UMA-style MiniMax M2.5 comparator produced similar aggregate and per-field classification metrics. Although predictive performance was similar, nMAS maintained unified report-level outputs with supporting source sentences; the demonstrated contribution is therefore workflow integration and traceability rather than predictive superiority. Under an illustrative, unmeasured scenario, reviewing 1,000 reports at five minutes per manual review versus five seconds per evidence-linked verification would reduce review time from 83.3 to 1.4 staff-hours, corresponding to 81.9 staff-hours and about USD~6,100 in potential staff-time value. Larger multi-institutional studies should evaluate evidence-span correctness, clinician verification time, and generalizability.

Yufan Wang, Anit Kumar Sahu, Yan Fei Ng +8

Nimblemind.ai, USA · Independent Researcher · Department of Anatomical Pathology, Singapore General Hospital, Singapore +2

Computational PathologyClinical Information ExtractionMulti-Agent Systems
AskPDF
Jul 7, 2026

A framework for single and multi-agent human-AI curiosity ecosystems

This paper offers a framework for considering curiosity as an ecosystem. First, it suggests that a single agent's inquiry policy (how, when, and why an agent asks a question) depends on how the agent values immediate uncertainty reduction, costs, delayed return, and the value of keeping the question open. A key concept in the framework is that the weights on these decision-related terms can change with experience. For example, a period of cheap, quickly answered questions may change the cost of inquiry on a short timescale and change which kinds of questions the agent is drawn to answer over a longer timescale. Second, these ideas are extended to many agents exploring a shared knowledge landscape, and there the framework tracks inquiry volume, topic diversity, frontier-directed inquiry, redundancy, and reusable knowledge. The result is a conceptual framework for studying curiosity ecology and for future efforts towards designing multi-agent AI systems for discovery.

Ilya E. Monosov

Solomon H. Snyder Department of Neuroscience, Johns Hopkins University, Baltimore, MD, USA · Departments of Biomedical Engineering, Electrical and Computer Engineering, and Psychiatry, Johns Hopkins University, Baltimore, MD, USA · Zanvyl Krieger Mind/Brain Institute, Johns Hopkins University, Baltimore, MD, USA +1

AI Agents for Scientific DiscoveryIntrinsic MotivationMulti-Agent Systems
AskPDF
Jul 7, 2026

Norm Enforcement for AI Agents: Robustly Shaping Behavior in Multi-Agent Systems

AI agents are increasingly deployed in shared environments where they pursue diverse goals and compete for rewards. This multi-agent competition can lead to behaviors that serve individual gains at collective cost -- for instance, marketing agents may post misleading content as a result of competing for engagement on social media. Human societies address such problems through norms that constrain acceptable behavior, supported by enforcement mechanisms that detect and penalize violations. Motivated by this, we study norm enforcement mechanisms for language model agents. We find that simple enforcement mechanisms are exploited by misaligned agents for competitive advantage, even when they are not explicitly trained or prompted to do so. We thus turn our attention to designing more robust mechanisms, and identify two key ingredients: estimating each agent's reliability over time, and updating this estimate with escalating penalties for repeated misbehavior. Across three simulated environments and a variety of agent populations, mechanisms built on these principles resist exploitation, while still penalizing norm violations at comparable or lower cost than baselines. Our results position norm enforcement mechanisms as scalable levers for shaping agents' behavior, but only when designed to anticipate becoming part of the system they govern. Our code and data are available at https://yaowenye.com/norm-enforcement.

Yaowen Ye, Jacob Steinhardt

University of California, Berkeley

LLM AgentsMulti-Agent SystemsAI Agent Governance
AskPDF
Jul 7, 2026

StateFuse: Deterministic Conflict-Preserving Memory for Multi-Agent Systems

Agent systems accumulate conflicting observations across branches, retries, and replicas, yet many practical memory layers still collapse disagreement behind overwrite rules that are difficult to inspect or correct. We present StateFuse, a conflict-aware replicated memory contract built on standard OpSet/CRDT merge. StateFuse does not introduce a new join algebra; it defines an agent-facing semantics layer with immutable history, explicit conflict objects, exact and semantic correction handles (claim_id / claim_ref), deterministic predicate contracts, and projection-time resolution that cannot rewrite replicated state. We evaluate StateFuse against flat multi-value, raw-log, provenance-style, and collapsed baselines under matched resolver and verification policies. On a 282-question official conflict-bearing MemoryAgentBench slice, the compared methods tie on answer accuracy, but conflict-preserving surfaces keep contradictions visible while collapsed surfaces do not. In a controlled agent loop with uniform verification, preserving ambiguity enables safer abstention and correction than early collapse. A correction-handle ablation further shows that semantic handles matter when exact prior identifiers are unavailable. The resulting claim is narrow: StateFuse is best supported as a safer public memory contract for contradiction surfacing, abstention, and auditable correction, not as a universal accuracy gain.

Sergey Volkov, Yang Li, Ye Luo

The University of Hong Kong

Agent MemoryMulti-Agent SystemsAgent Memory ManagementConsistency Models
AskPDF
Jul 6, 2026

Collective Intelligence with Foundation Models

As foundation models grow in scale and diversity, coordinating multiple models into cooperative reasoning systems offers a path toward safer, more reliable AI. This chapter presents a multi-agent framework where solver models generate independent drafts, each undergoes structured critique and revision by a critic agent, and an aggregator agent synthesizes a final consensus solution. A scoring module provides semantic, numerical, and procedural evaluation across all agents. Through ablation studies on a benchmark spanning calculus, physics, chemistry, biology, economics, optimization, statistics, and mathematics, we isolate the contributions of framework architecture versus model diversity. We compare four configurations: (1) Individual Baseline, (2) Homogeneous Framework using one shared model, (3) Redundant Homogeneous Solvers using multiple instances of the same model, and (4) Heterogeneous Framework with diverse specialized models. Results show that while framework structure and redundant sampling yield modest gains, model heterogeneity is the critical factor driving substantial performance improvements. The heterogeneous configuration achieves superior step-wise accuracy (0.64 vs. 0.54 for individual models; 2.3x improvement over homogeneous configurations) with reduced variance across categories and difficulty levels. Step-wise reasoning quality (correctness of intermediate steps, not just final answers) improves dramatically only with model diversity, showing that heterogeneous agents provide complementary error detection and reasoning refinement essential for explainability and auditability. We discuss architectural principles, evaluation methodology, and implications for Global Applied AI, showing how heterogeneous multi-agent coordination supports transparent, auditable, high-confidence decision-making across scientific and industrial domains.

J. de Curtò, I. de Zarzà

Department of Computer Applications in Science & Engineering, BARCELONA Supercomputing Center, 08034 Barcelona, Spain · Escuela Técnica Superior de Ingeniería (ICAI), Universidad Pontificia Comillas, 28015 Madrid, Spain · Human Centered AI, Data & Software, LUXEMBOURG Institute of Science and Technology, L-4362 Esch-sur-Alzette, Luxembourg

AI Agent EvaluationMulti-Agent CollaborationMulti-Agent SystemsMulti-Agent ConsensusLanguage Model EnsemblesMulti-Agent Reasoning
AskPDF
Jul 6, 2026

Decision Protocols in Multi-Agent Large Language Model Conversations

Improving the task performance of Large Language Models (LLMs) is essential, yet scaling these models faces significant challenges such as diminishing returns and high costs. Multi-Agent Systems (MAS) offer a promising solution by distributing tasks among specialized agents to improve the overall task performance. This can reduce training costs at the expense of increased test time due to the discussion and decision-making process. The decision protocol is a critical component of MAS because it specifies how multiple agents collaborate to create a final solution. This thesis introduces the Multi-Agent LLM (MALLM) framework, which implements and evaluates various decision protocols, namely voting, consensus, and judge decision mechanisms, to simulate multi-agent discussions for conversational task solving. Unlike previous work that used a single decision protocol or tested them on limited datasets, this study systematically examines their impact on a diverse set of tasks, ranging from knowledge-based datasets (MMLU, MMLU-Pro, GPQA) and logic-based datasets (StrategyQA, MuSR, Math-lvl-5, SQuAD 2.0). The results indicate that consensus protocols excel in knowledge-intensive domains while voting and judge protocols are more effective for logic-based tasks. Increasing response diversity through independent solution generation improves decision quality, while changes in information access during the decision process have minimal impact.

Lars Benedikt Kaesberg

GippLab Georg August University of Göttingen Göttingen

Multi-Agent LLM SystemsLLM Agent EvaluationMulti-Agent CollaborationMulti-Agent SystemsMulti-Agent Reasoning
AskPDF
Jul 6, 2026

An Exploration of Agentic Information Fusion for Test Maintenance Prediction

Test maintenance is a critical, yet costly, activity - particularly as codebases rapidly evolve. To assist, we present MAST, a multi-agent framework that predicts which test cases require maintenance following changes to the production code. This identification task is necessary as a precondition to any subsequent maintenance activities, but remains challenging due to the complex relationships between production and test code. MAST advances the state-of-the-art by integrating multiple analyses -- including static, lexical, and semantic analyses - through an intelligent fusion and post-check procedure and by focusing on a realistic use and evaluation setting - i.e., standardized input formats, repository-level analyses, and the ability to infer relations between test and production artifacts rather than assuming a pre-existing mapping. We evaluated MAST on 21 industrial Java repositories from Ericsson AB, considering situations where test maintenance both was and was not required in the ground truth. MAST yielded superior precision to a state-of-the-art baseline - resulting in a higher accuracy, F1, and F2 score - with only some loss in recall. Our ablation study demonstrates the value of each analysis in producing the final recommendations. MAST illustrates the potential of multi-agent systems that can fuse multiple information sources when performing software testing tasks.

Jingxiong Liu, Nasser Mohammadiha, Gregory Gay

Chalmers University of Technology and University of Gothenburg · Ericsson AB

Automated Software TestingPredictive MaintenanceMulti-Agent Systems
AskPDF
Jul 3, 2026

Replicating Belief, Not Bits: Epistemic State Replication for Agentic Systems

In distributed systems, the classical State Machine Replication (SMR) model assumes that correct replicas execute deterministic transitions to yield identical bitwise states. However, the rise of agentic distributed systems -- where autonomous, stochastic, and model-driven agents orchestrate infrastructure -- presents scenarios where deterministic, bitwise replication is insufficient. Replicas operating with generative models may exhibit divergent reasoning paths, summaries, and token boundaries, yet reach semantically equivalent and correct operational decisions. Forcing bitwise agreement across these stochastic participants degrades execution flexibility, induces context amnesia, and limits performance. We argue that in such settings replicas should agree on belief, not bits. We propose Epistemic State Replication (ESR), a belief-replication layer for agentic distributed systems that shifts the replication boundary from data visibility to knowledge visibility. We formalize the epistemic node state as a pair K = (L, B) separating the deterministic, immutable evidence log (L) from the stochastic, evolving belief lineage (B). To govern execution safety, we define Semantic Linearizability, which requires operations to reflect the latest committed operational meaning within a verifier-bounded semantic compatibility metric, and Bounded Eventual Coherence, which bounds expected semantic divergence under fair delivery, monotonic evidence, bounded verifier disturbance, and a contractive graft operator. We outline protocols for propagating derived insights using structured epistemic deltas, and formalize Verifiable Semantic Rollbacks to prune faulty premises from belief lineages without inducing context amnesia. We prototype ESR and report preliminary simulation results that show feasibility under the stated assumptions and illustrate reductions in secondary cognitive faults.

Jun He, Deying Yu

Multi-Agent Systems
AskPDF
Jul 3, 2026

Second MOASEI Competition at AAMAS'2026: A Technical Report

We describe the 2026 Methods for Open Agent Systems Evaluation Initiative (MOASEI) Competition, a benchmark event for evaluating multi-agent decision-making under open-system conditions. Building on the inaugural 2025 competition, the 2026 edition retained wildfire fighting, cybersecurity, and ride-sharing domains while adding a bonus wildfire track with frame openness, in which agent equipment states such as suppressant capacities and firefighting range vary over time. The competition also expanded its reporting metrics to emphasize total task completions, mean task-completion time, and mean value of completed tasks. Participation in 2026 was limited: eight teams registered, but only one team submitted a final entry, and that entry targeted the ride-sharing track. The submitted DLC approach used planning and replanning to solve routing problems across agents as passengers appeared. This report summarizes the 2026 competition design, highlights differences from the previous year, and reports ride-sharing evaluation results against baseline policies. DLC is recognized as the 2026 ride-sharing track winner among submitted teams.

Ceferino Patino, Tyler J. Billings, Alireza Saleh Abadi +4

University of Nebraska - Lincoln · Lincoln, Nebraska, United States · University of Georgia - Athens +3

Multi-Agent CoordinationAI Agent BenchmarksMulti-Agent SystemsMulti-Agent Planning
AskPDF