Multi-Agent LLM Systems

LLM: Large Language Model

Ask about this topic

Overview

Multi-agent large language model systems study how multiple language-model agents can divide work, share information, and combine decisions on tasks such as reasoning, coding, research, and planning. Common designs use specialized roles, debate or review loops, retrieval-augmented generation, shared memory, and orchestrators that route tasks or choose communication links; some systems also learn prompts, workflows, or policies through reinforcement learning. Researchers evaluate these approaches on benchmarks for math, question answering, software, and tool use, as well as in longer-running simulations and domain applications. A central open tension is that collaboration can improve accuracy and verification, but may also add latency and token costs or spread errors, bias, and unsafe instructions, so gains depend on the task, architecture, and quality of coordination.

Latest papers

CardsList

1,050 papers142 papers in the last four weeks, up 35% on the four weeks before. 0.8% of all new papers.

Apr 16, 2026

El Agente Forjador: Task-Driven Agent Generation for Quantum Simulation

AI for science promises to accelerate the discovery process. The advent of large language models (LLMs) and agentic workflows enables the expediting of a growing range of scientific tasks. However, most of the current generation of agentic systems depend on static, hand-curated toolsets that hinder adaptation to new domains and evolving libraries. We present El Agente Forjador, a multi-agent framework in which universal coding agents autonomously forge, validate, and reuse computational tools through a four-stage workflow of tool analysis, tool generation, task execution, and iterative solution evaluation. Evaluated across 24 tasks spanning quantum chemistry and quantum dynamics on five coding agent setups, we compare three operating modes: zero-shot generation of tools per task, reuse of a curriculum-built toolset, and direct problem-solving with the coding agents as the baseline. We find that our tool generation and reuse framework consistently improves accuracy over the baseline. We also show that reusing a toolset built by a stronger coding agent can reduce API cost and substantially raises the solution quality for weaker coding agents. Case studies further demonstrate that tools forged for different domains can be combined to solve hybrid tasks. Taken together, these results show that LLM-based agents can use their scientific knowledge and coding capabilities to autonomously build reusable scientific tools, pointing toward a paradigm in which agent capabilities are defined by the tasks they are designed to solve rather than by explicitly engineered implementations.

Zijian Zhang, Aiwei Yin, Amaan Baweja +4

Department of Computer Science, University of Toronto, 40 St George St., Toronto, ON M5S 2E4, Canada · Vector Institute for Artificial Intelligence, W1140-108 College St., Schwartz Reisman Innovation2026 Campus, Toronto, ON M5G 0C6, Canada · NVIDIA, 431 King St W #6th, Toronto, ON M5V 1K4, Canada +5

Multi-Agent LLM SystemsAI for ScienceScientific Code GenerationTool-Using AgentsLLM AgentsLLM Agent Tool Generation
AskPDF
Apr 16, 2026

Mind DeepResearch Technical Report

We present Mind DeepResearch (MindDR), an efficient multi-agent deep research framework that achieves leading performance with only ~30B-parameter models through a meticulously designed data synthesis and multi-stage training pipeline. The core innovation of MindDR lies in a collaborative three-agent architecture (Planning Agent, DeepSearch Agent, and Report Agent) and a four-stage agent-specialized training pipeline comprising SFT cold-start, Search-RL, Report-RL and preference alignment. With this regime, MindDR demonstrates competitive performance even with ~30B-scale models. Specifically, MindDR achieves 45.7% on BrowseComp-ZH, 42.8% on BrowseComp, 46.5% on WideSearch, 75.0% on xbench-DS, and 52.5 on DeepResearch Bench, outperforming comparable-scale open-source agent systems and rivaling larger-scale models. MindDR has been deployed as an online product in Li Auto. Furthermore, we introduce MindDR Bench, a curated benchmark of 500 real-world Chinese queries from our internal product user interactions, evaluated through a comprehensive multi-dimensional rubric system rather than relying on a single RACE metric. On MindDR Bench, MindDR achieves a state-of-the-art score of 51.8.

MindDR Team, Li Auto Inc

MindDR Team, Li Auto Inc

Multi-Agent LLM SystemsDeep Research AgentsAI Agent BenchmarksLLM Agent TrainingRL for LLM Agents
AskPDF
Apr 15, 2026

Coalition Formation in LLM Agent Networks: Stability Analysis and Convergence Guarantees

Large Language Model (LLM) agents are increasingly deployed in multi-agent systems requiring strategic coordination. While recent work has analyzed LLM behavior in two-player games, coalition formation, where nn agents dynamically form cooperative groups, remains theoretically uncharacterized. We present the first framework grounding coalition formation in LLM agent networks in hedonic game theory with formal stability guarantees. We introduce the LLM Coalition Formation Game (LCFG), establish sufficient conditions for Nash-stable partitions, and prove complexity results. Our analysis reveals that LLM agents exhibit bounded rationality characterized by εε-rational preferences; we provide both deterministic existence guarantees and consistency-driven stability bounds whose predictions are consistent with empirical outcomes. Experiments with GPT-4, Claude-3, and Llama-3 across 2,400 episodes validate our framework: LLM coalitions achieve Nash stability in 73.2% of cases under our Coalition-of-Thought (CoalT) protocol, compared to 58.4% under chain-of-thought and 41.8% under standard prompting (p<0.001p < 0.001). Our framework provides theoretical foundations for designing stable multi-agent LLM systems.

Dongxin Guo, Jikun Wu, Siu-Ming Yiu

Department of Compute Science, The University of Hong Kong, Hong Kong, China · Brain Investing Limited, Hong Kong, China · Stellaris AI Limited, Hong Kong, China

Coalition FormationMulti-Agent LLM SystemsCooperative Game Theory
AskPDF
Apr 15, 2026

TREX: Automating LLM Fine-tuning via Agent-Driven Tree-based Exploration

While Large Language Models (LLMs) have empowered AI research agents to perform isolated scientific tasks, automating complex, real-world workflows, such as LLM training, remains a significant challenge. In this paper, we introduce TREX, a multi-agent system that automates the entire LLM training life-cycle. By orchestrating collaboration between two core modules-the Researcher and the Executor-the system seamlessly performs requirement analysis, open-domain literature and data research, formulation of training strategies, preparation of data recipes, and model training and evaluation. The multi-round experimental process is modeled as a search tree, enabling the system to efficiently plan exploration paths, reuse historical results, and distill high-level insights from iterative trials. To evaluate the capability of automated LLM training, we construct FT-Bench, a benchmark comprising 10 tasks derived from real-world scenarios, ranging from optimizing fundamental model capabilities to enhancing performance on domain-specific tasks. Experimental results demonstrate that the TREX agent consistently optimizes model performance on target tasks.

Zerun Ma, Guoqiang Wang, Xinchen Xie +7

Shanghai AI Laboratory · Fudan University

LLM Fine-TuningLLM Agent TrainingMulti-Agent LLM Systems
AskPDF
Apr 15, 2026

ReviewGrounder: Improving Review Substantiveness with Rubric-Guided, Tool-Integrated Agents

The rapid rise in AI conference submissions has driven increasing exploration of large language models (LLMs) for peer review support. However, LLM-based reviewers often generate superficial, formulaic comments lacking substantive, evidence-grounded feedback. We attribute this to the underutilization of two key components of human reviewing: explicit rubrics and contextual grounding in existing work. To address this, we introduce REVIEWBENCH, a benchmark evaluating review text according to paper-specific rubrics derived from official guidelines, the paper's content, and human-written reviews. We further propose REVIEWGROUNDER, a rubric-guided, tool-integrated multi-agent framework that decomposes reviewing into drafting and grounding stages, enriching shallow drafts via targeted evidence consolidation. Experiments on REVIEWBENCH show that REVIEWGROUNDER, using a Phi-4-14B-based drafter and a GPT-OSS-120B-based grounding stage, consistently outperforms baselines with substantially stronger/larger backbones (e.g., GPT-4.1 and DeepSeek-R1-670B) in both alignment with human judgments and rubric-based review quality across 8 dimensions. The code is available here.

Zhuofeng Li, Yi Lu, Dongfu Jiang +7

Texas A&M University · University of Waterloo · UC San Diego +2

Automated Peer ReviewScholarly Peer ReviewRubric-Based EvaluationMulti-Agent LLM Systems
AskPDF
Apr 15, 2026

MIND: AI Co-Scientist for Material Research

Large language models (LLMs) have enabled agentic AI systems for scientific discovery, but most approaches remain limited to textbased reasoning without automated experimental verification. We propose MIND, an LLM-driven framework for automated hypothesis validation in materials research. MIND organizes the scientific discovery process into hypothesis refinement, experimentation, and debate-based validation within a multi-agent pipeline. For experimental verification, the system integrates Machine Learning Interatomic Potentials, particularly SevenNet-Omni, enabling scalable in-silico experiments. We also provide a web-based user interface for automated hypothesis testing. The modular design allows additional experimental modules to be integrated, making the framework adaptable to broader scientific workflows. The code is available at: https://github.com/IMMS-Ewha/MIND, and a demonstration video at: https://youtu.be/lqiFe1OQzN4.

Geonhee Ahn, Donghyun Lee, Hayoung Doo +3

Institute for Multiscale Matter and Systems (IMMS), Ewha Womans University, Seoul 03760, Republic of Korea

AI Agents for Scientific DiscoveryAutonomous ExperimentationMaterials ScienceMulti-Agent LLM Systems
AskPDF
Apr 14, 2026

Toward Autonomous Long-Horizon Engineering for ML Research

Agentic systems increasingly automate pieces of AI research. Yet turning underspecified research objectives into runnable, experimentally validated ML systems remains a central bottleneck. We study this operational setting as long-horizon ML research engineering: converting a research specification into a runnable ML system through repeated implementation, experimentation, and refinement. The central challenge is to sustain cumulative project progress across heterogeneous stages under delayed, confounded feedback. We introduce AiScientist, a multi-agent system built around thin control over thick state: a lightweight hierarchical research team coordinates through a File-as-Bus workspace that preserves decision-relevant artifacts across roles and invocations. On PaperBench, AiScientist improves over the strongest matched baselines by 9.92 and 11.15 points with Gemini-3-Flash and GLM-5, respectively. On MLE-Bench Lite, it reaches 81.82 Any Medal% under both backbones, improving over the strongest matched baselines by 4.55 and 16.67 points, and exceeding a Codex/GPT-5.5 xhigh frontier harness reference by 13.64 Any Medal points. Ablations and process analyses show that durable project state is central to later-round refinement: removing File-as-Bus lowers PaperBench score by 6.41 points and MLE-Bench Lite Any Medal% by 31.82 points. These results suggest that long-horizon AI research is not only a problem of stronger local reasoning, but a systems problem of maintaining cumulative, inspectable project progress.

Guoxin Chen, Jie Chen, Lei Chen +7

Long-Horizon LLM AgentsMulti-Agent LLM SystemsMemory-Augmented Agents
AskPDF
Apr 14, 2026

TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanations

Traditional conversational travel recommender systems primarily optimize for user relevance and convenience, often reinforcing popular, overcrowded destinations and carbon-intensive travel choices. To address this, we present TRACE (Tourism Recommendation with Agentic Counterfactual Explanations), a multi-agent, LLM-based framework that promotes sustainable tourism through interactive nudging. TRACE uses a modular orchestrator-worker architecture where specialized agents elicit latent sustainability preferences, construct structured user personas, and generate recommendations that balance relevance with environmental impact. A key innovation lies in its use of agentic counterfactual explanations and LLM-driven clarifying questions, which together surface greener alternatives and refine understanding of intent, fostering user reflection without coercion. User studies and semantic alignment analyses demonstrate that TRACE effectively supports sustainable decision-making while preserving recommendation quality and interactive responsiveness. TRACE is implemented on Google's Agent Development Kit, with full code, Docker setup, prompts, and a publicly available demo video to ensure reproducibility. A project summary, including all resources, prompts, and demo access, is available at https://ashmibanerjee.github.io/trace-chatbot.

Ashmi Banerjee, Adithi Satish, Wolfgang Wörndl +1

Technical University of Munich, Munich, Germany · Polytechnic University of Bari, Bari, Italy

Large Language Model-Based RecommendationExplainable RecommendationMulti-Agent LLM Systems
AskPDF
Apr 14, 2026

CIA: Inferring the Communication Topology from LLM-based Multi-Agent Systems

LLM-based Multi-Agent Systems (MAS) have demonstrated remarkable capabilities in solving complex tasks. Central to MAS is the communication topology which governs how agents exchange information internally. Consequently, the security of communication topologies has attracted increasing attention. In this paper, we investigate a critical privacy risk: MAS communication topologies can be inferred under a restrictive black-box setting, exposing system vulnerabilities and posing significant intellectual property threats. To explore this risk, we propose Communication Inference Attack (CIA), a novel attack that constructs new adversarial queries to induce intermediate agents' reasoning outputs and models their semantic correlations through the proposed global bias disentanglement and LLM-guided weak supervision. Extensive experiments on MAS with optimized communication topologies demonstrate the effectiveness of CIA, achieving an average AUC of 0.87 and a peak AUC of up to 0.99, thereby revealing the substantial privacy risk in MAS.

Yongxuan Wu, Xixun Lin, He Zhang +5

Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China · School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China · Griffith University, Brisbane, Australia +2

Multi-Agent System SecurityMulti-Agent LLM SystemsMulti-Agent Communication
AskPDF
Apr 13, 2026

Mathematics Teachers Interactions with a Multi-Agent System for Personalized Problem Generation

Large language models can increasingly adapt educational tasks to learners characteristics. In the present study, we examine a multi-agent teacher-in-the-loop system for personalizing middle school math problems. The teacher enters a base problem and desired topic, the LLM generates the problem, and then four AI agents evaluate the problem using criteria that each specializes in (mathematical accuracy, authenticity, readability, and realism). Eight middle school mathematics teachers created 212 problems in ASSISTments using the system and assigned these problems to their students. We find that both teachers and students wanted to modify the fine-grained personalized elements of the real-world context of the problems, signaling issues with authenticity and fit. Although the agents detected many issues with realism as the problems were being written, there were few realism issues noted by teachers and students in the final versions. Issues with readability and mathematical hallucinations were also somewhat rare. Implications for multi-agent systems for personalization that support teacher control are given.

Candace Walkington, Theodora Beauchamp, Fareya Ikram +4

Southern Methodist University, Dallas TX 75205, USA · University of Massachusetts Amherst, Amherst MA 01003, USA · Worchester Polytechnic Institute, Worcester, MA 01609 USA

Mathematical Problem GenerationGenerative AI in EducationMulti-Agent LLM SystemsMathematics Education
AskPDF
Apr 13, 2026

M2HRI: An LLM-Driven Multimodal Multi-Agent Framework for Personalized Human-Robot Interaction

Multi-robot systems hold significant promise for social environments such as homes and hospitals, yet existing multi-robot systems often treat robots as functionally interchangeable, overlooking how distinct agent identities shape user perception and how such individuality changes the coordination requirements of multi-robot interaction. To address this, we introduce M2HRI, a multimodal multi-agent framework that models each robot as an identity-bearing agent through personality and long-term memory, together with a contextualized coordination mechanism that regulates agent participation. In a controlled user study (n = 105) in a multi-agent human-robot interaction (HRI) scenario, we found that most personality contrasts were distinguishable and consistently expressed. Long-term memory improved preference awareness and interaction naturalness, while contextualized coordination improved conversational flow, response appropriateness, and overlap avoidance. Together, these findings show that agent individuality and contextualized participation coordination play complementary roles in supporting coherent and socially appropriate multi-agent HRI. Project website available at https://project-m2hri.github.io/.

Shaid Hasan, Breenice Lee, Sujan Sarker +1

University of Virginia, VA, USA

Multi-Agent LLM SystemsMulti-Agent CoordinationHuman-Robot InteractionMultimodal AgentsMemory-Augmented Agents
AskPDF
Apr 13, 2026

Collaborative Multi-Agent Scripts Generation for Enhancing Imperfect-Information Reasoning in Murder Mystery Games

Vision-language models (VLMs) have shown impressive capabilities in perceptual tasks, yet they degrade in complex multi-hop reasoning under multiplayer game settings with imperfect and deceptive information. In this paper, we study a representative multiplayer task, Murder Mystery Games, which require inferring hidden truths based on partial clues provided by roles with different intentions. To address this challenge, we propose a collaborative multi-agent framework for evaluating and synthesizing high-quality, role-driven multiplayer game scripts, enabling fine-grained interaction patterns tailored to character identities (i.e., murderer vs. innocent). Our system generates rich multimodal contexts, including character backstories, visual and textual clues, and multi-hop reasoning chains, through coordinated agent interactions. We design a two-stage agent-monitored training strategy to enhance the reasoning ability of VLMs: (1) chain-of-thought based fine-tuning on curated and synthetic datasets that model uncertainty and deception; (2) GRPO-based reinforcement learning with agent-monitored reward shaping, encouraging the model to develop character-specific reasoning behaviors and effective multimodal multi-hop inference. Extensive experiments demonstrate that our method significantly boosts the performance of VLMs in narrative reasoning, hidden fact extraction, and deception-resilient understanding. Our contributions offer a scalable solution for training and evaluating VLMs under uncertain, adversarial, and socially complex conditions, laying the groundwork for future benchmarks in multimodal multi-hop reasoning under imperfect information.

Keyang Zhong, Junlin Xie, Hefeng Wu +2

Sun Yat-sen University · Shenzhen Loop Area Institute · The Chinese University of Hong Kong, Shenzhen

VLM ReasoningMultimodal ReasoningMulti-Agent LLM SystemsRL for Language Model ReasoningLarge Language Model-Based Role-Play Simulation
AskPDF
Apr 13, 2026

A collaborative agent with two lightweight synergistic models for autonomous crystal materials research

Current large language models require hundreds of billions of parameters yet struggle with domain-specific reasoning and tool coordination in materials science. Here, we present MatBrain, a lightweight collaborative agent system with two synergistic models specialization for crystal materials research. MatBrain employs a dual-model architecture: Mat-R1 (30B parameters) as the analytical model providing expert-level domain reasoning, and Mat-T1 (14B parameters) as the executive model orchestrating tool-based actions. Entropy analysis confirms that this architecture resolves the conflict between tool planning and analytical reasoning by decoupling their distinct entropy dynamics. Enabled by this dual-model architecture and structural efficiency, MatBrain significantly outperforms larger general-purpose models while reducing the hardware deployment barrier by over 95%. MatBrain exhibits versatility across structure generation, property prediction, and synthesis planning tasks. Applied to catalyst design, MatBrain generated 30,000 candidate structures and identified 38 promising materials within 48 hours, achieving approximately 100-fold acceleration over traditional approaches. These results demonstrate the potential of lightweight collaborative intelligence for advancing materials research capabilities.

Tongyu Shi, Yutang Li, Zhanyuan Li +9

Materials Artificial Intelligence Center, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, 518055, China · University of Chinese Academy of Sciences, Beijing 100049, China · Institute for Advanced Study, Chengdu University, Chengdu 610106, Sichuan, China. +1

Autonomous Research AgentsMulti-Agent LLM SystemsMaterials ScienceCrystal Structure GenerationMaterials Property Prediction
AskPDF
Apr 13, 2026

Collective Opinion Dynamics in Structured LLM Populations

Large Language Models are increasingly deployed as interacting agents in settings such as online platforms, recommendation systems, and multi-agent applications. Understanding the collective behaviors that emerge from their interactions is therefore increasingly crucial, especially as these behaviors may shape public opinion and contribute to polarization. In this work, we investigate how network structure and group composition shape the evolution of opinions in populations of LLM agents engaged in multi-round debates. We generate networks with controlled levels of homophily and varying group sizes, and perform ten independent runs per configuration. The results show that LLM agents exhibit patterns that are highly sensitive to network structure, relative group sizes, and to the model itself. We highlight that even within LLM populations, low homophily facilitates convergence by increasing opportunities for cross-group interaction, whereas higher homophily limits cross-group exposure and can preserve distinct opinion states, consistent with previous research. We further show that the choice of LLM substantially affects the resulting dynamics, with different models exhibiting different patterns of opinion updating under comparable network conditions. At the individual level, providing agents with information about their local neighborhood further modulates opinion transitions, revealing substantial differences between models in their sensitivity to local social context. Overall, our findings show that the collective dynamics of LLM populations arise from interactions among model-specific behavior, network structure, population composition, and local social information.

Erica Cau, Andrea Failla, Giulio Rossetti

Department of Computer Science, University of Pisa · ISTI-CNR

Multi-Agent LLM SystemsOpinion DynamicsMulti-Agent DebateAgent-Based Modeling
AskPDF
Apr 13, 2026

ArtiCAD: Articulated CAD Assembly Design via Multi-Agent Code Generation

Parametric Computer-Aided Design (CAD) of articulated assemblies is essential for product development, yet generating these multi-part, movable models from high-level descriptions remains unexplored. To address this, we propose ArtiCAD, the first training-free multi-agent system capable of generating editable, articulated CAD assemblies directly from text or images. Our system divides this complex task among four specialized agents: Design, Generation, Assembly, and Review. One of our key insights is to predict assembly relationships during the initial design stage rather than the assembly stage. By utilizing a Connector that explicitly defines attachment points and joint parameters, ArtiCAD determines these relationships before geometry generation, effectively bypassing the limited spatial reasoning capabilities of current LLMs and VLMs. To further ensure high-quality outputs, we introduce validation steps in the generation and assembly stages, accompanied by a cross-stage rollback mechanism that accurately isolates and corrects design- and code-level errors. Additionally, a self-evolving experience store accumulates design knowledge to continuously improve performance on future tasks. Extensive evaluations on three datasets (ArtiCAD-Bench, CADPrompt, and ACD) validate the effectiveness of our approach. We further demonstrate the applicability of ArtiCAD in requirement-driven conceptual design, physical prototyping, and the generation of embodied AI training assets through URDF export.

Yuan Shui, Yandong Guan, Zhanwei Zhang +4

School of Software, Beihang University · Zhejiang University · The University of Hong Kong

Text-to-CAD GenerationParametric CAD ModelingMulti-Agent LLM SystemsLLM-Based Program Synthesis
AskPDF
Apr 13, 2026

MAFIG: Multi-agent Driven Formal Instruction Generation Framework

Emergency situations in scheduling systems often trigger local functional failures that undermine system stability and even cause system collapse. Existing methods primarily rely on robust scheduling or reactive scheduling, handling emergencies through predefined rules or rescheduling strategies. However, the diversity and unpredictability of real-world emergencies make them difficult to anticipate, which limits the adaptability of these methods in complex scenarios. Recent studies have shown that Large Language Models (LLMs) possess strong potential for complex scheduling tasks because of their extensive prior knowledge and strong reasoning capabilities. Nevertheless, the high inference latency of LLMs and the lengthy contextual information of scheduling systems significantly hinder their application for emergency handling. To mitigate these issues, we propose the Multi-agent Driven Formal Instruction Generation Framework (MAFIG). The framework constrains the decision scope to local functional modules affected by emergency situations and repairs scheduling logic rapidly by generating formal instructions. MAFIG contains a Perception Agent and an Emergency Decision Agent, which mitigates the adverse impact of lengthy system contexts on emergency decision-making. We further introduce span-focused loss-driven local distillation mechanism (SFL) to transfer the decision-making capability of powerful Cloud Large Language Models (C-LLMs) to lightweight local models, reducing inference latency while preserving decision-making effectiveness. Experiments in the Port, Warehousing, and Deck scheduling datasets show success rates of 98.49%, 94.97%, and 97.50%, with average processing times of 0.33 s, 0.23 s, and 0.19 s. These results demonstrate that MAFIG effectively mitigates the impact of emergencies and improves the robustness and adaptability of scheduling systems.

Shixing Zhao, Zheng Si, Pengpeng Ouyang +5

School of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou, 450001, Henan, China · Engineering Research Center of Intelligent Swarm Systems, Ministry of Education, Zhengzhou, 450001, Henan, China · National Supercomputing Center in Zhengzhou, Zhengzhou, 450001, Henan, China +1

Synthetic Instruction GenerationMulti-Agent LLM Systems
AskPDF
Apr 12, 2026

Diversity Without Fidelity: A Solver-Sampler Mismatch in Multi-Agent LLM Negotiation Simulation

Language models are increasingly used to simulate people: survey respondents, negotiators, stakeholders in policy exercises. In that role a model should reproduce how people plausibly behave, hesitating, conceding late, and settling for imperfect deals, rather than playing the best move. We call this the sampler role, in contrast to the solver role of finding the best move, and we test how the reasoning modes providers ship to strengthen models as solvers affect it. Our testbed is multi-party negotiation: five agents bargain over a regulation for fifteen turns, and unresolved issues are decided by an authority. Agents without a structured memory of the negotiation almost never reach agreement, whether reasoning is on or off: 314 of 315 such runs end with the authority deciding. What reasoning changes is how the failure looks. With reasoning enabled, one model family negotiates visibly, with varied moves, concessions in most runs, and a different path every time, yet still ends without agreement in fifteen runs of fifteen. Diversity checks would pass a model whose endings never change. Two further results show the task permits agreement: when agents write their own short running notes on the negotiation, agreement becomes the norm, while the same notes supplied ready-made change nothing; and hand-coded agents following textbook concession strategies agree in most runs under identical rules. Simulation pipelines should therefore vet models as samplers, on the distributions of outcomes they produce. Fidelity as a sampler must be tested on its own: solver strength is no guide to it, and switching on reasoning leaves it where it was.

Sandro Andric

Multi-Agent LLM SystemsMulti-Agent NegotiationAutomated NegotiationLLM Agent EvaluationLarge Language Model-Based Social Simulation
AskPDF
Apr 12, 2026

From Query to Counsel: Structured Reasoning with a Multi-Agent Framework and Dataset for Legal Consultation

Legal consultation question answering (Legal CQA) presents unique challenges compared to traditional legal QA tasks, including the scarcity of high-quality training data, complex task composition, and strong contextual dependencies. To address these, we construct JurisCQAD, a large-scale dataset of over 43,000 real-world Chinese legal queries annotated with expert-validated positive and negative responses, and design a structured task decomposition that converts each query into a legal element graph integrating entities, events, intents, and legal issues. We further propose JurisMA, a modular multi-agent framework supporting dynamic routing, statutory grounding, and stylistic optimization. Combined with the element graph, the framework enables strong context-aware reasoning, effectively capturing dependencies across legal facts, norms, and procedural logic. Trained on JurisCQAD and evaluated on a refined LawBench, our system significantly outperforms both general-purpose and legal-domain LLMs across multiple lexical and semantic metrics, demonstrating the benefits of interpretable decomposition and modular collaboration in Legal CQA.

Mingfei Lu, Yi Zhang, Mengjia Wu +1

Australian Artificial Intelligence Institute (AAII), University of Technology Sydney · School of Computer Science, University of Birmingham

Legal QALegal ReasoningMulti-Agent LLM SystemsLLM Reasoning with Graphs
AskPDF
Apr 12, 2026

TrajOnco: a multi-agent framework for temporal reasoning over longitudinal EHR for multi-cancer early detection

Accurate estimation of cancer risk from longitudinal electronic health records (EHRs) could support earlier detection and improved care, but modeling such complex patient trajectories remains challenging. We present TrajOnco, a training-free, multi-agent large language model (LLM) framework designed for scalable multi-cancer early detection. Using a chain-of-agents architecture with long-term memory, TrajOnco performs temporal reasoning over sequential clinical events to generate patient-level summaries, evidence-linked rationales, and predicted risk scores. We evaluated TrajOnco on de-identified Truveta EHR data across 15 cancer types using matched case-control cohorts, predicting risk of cancer diagnosis at 1 year. In zero-shot evaluation, TrajOnco achieved AUROCs of 0.64-0.80, performing comparably to supervised machine learning in a lung cancer benchmark while demonstrating better temporal reasoning than single-agent LLMs. The multi-agent design also enabled effective temporal reasoning with smaller-capacity models such as GPT-4.1-mini. The fidelity of TrajOnco's output was validated through human evaluation. Furthermore, TrajOnco's interpretable reasoning outputs can be aggregated to reveal population-level risk patterns that align with established clinical knowledge. These findings highlight the potential of multi-agent LLMs to execute interpretable temporal reasoning over longitudinal EHRs, advancing both scalable multi-cancer early detection and clinical insight generation.

Sihang Zeng, Young Won Kim, Wilson Lau +4

Truveta Inc., Bellevue, WA, United States. · University of Washington, Seattle, WA, United States. · Fred Hutch Cancer Center, Seattle, WA, United States.

Clinical Risk PredictionLongitudinal EHR ModelingMulti-Agent LLM SystemsTemporal Reasoning in Language ModelsMemory-Augmented Language Model Agents
AskPDF
Apr 10, 2026

Toward Explanatory Equilibrium: Verifiable Reasoning as a Coordination Mechanism under Asymmetric Information

LLM-based agents increasingly coordinate decisions in multi-agent systems, often attaching natural-language reasoning to actions. However, reasoning is neither free nor automatically reliable: it incurs computational cost and, without verification, may degenerate into persuasive cheap talk. We introduce Explanatory Equilibrium as a design principle for explanation-aware multi-agent systems and study a regime in which agents exchange structured reasoning artifacts-auditable claims paired with concise text-while receivers apply bounded verification through probabilistic audits under explicit resource constraints. We contribute (i) a minimal mechanism-level exchange-audit model linking audit intensity, misreporting incentives, and reasoning costs, and (ii) empirical evidence from a finance-inspired LLM setting involving a Trader and a Risk Manager. In ambiguous, borderline proposals, auditable artifacts prevent the cost of silence driven by conservative validation under asymmetric information: without structured claims, approval and welfare collapse. By contrast, structured reasoning unlocks coordination while maintaining consistently low bad-approval rates across audit intensities, audit budgets, and incentive regimes. Our results suggest that scalable, safety-preserving coordination in LLM-based multi-agent systems depends not only on audit strength, but more fundamentally on disciplined externalization of reasoning into partially verifiable artifacts.

Feliks Bańka, Jarosław A. Chudziak

The Faculty of Electronics and Information Technology, Warsaw University of Technology, Warsaw, Poland

Multi-Agent LLM SystemsMulti-Agent CoordinationReasoning VerificationMechanism DesignAI Agent Auditing
AskPDF
Apr 10, 2026

Enhancing LLM Problem Solving via Tutor-Student Multi-Agent Interaction

Human cognitive development is shaped not only by individual effort but by structured social interaction, where role-based exchanges such as those between a tutor and a learner, enable solutions that neither could achieve alone. Inspired by these developmental principles, we ask the question whether a tutor-student multi-agent system can create a synergistic effect by pushing Large Language Model (LLM) beyond what it can do within existing frameworks. To test the idea, we adopt autonomous coding problem domain where two agents instantiated from the same LLM assigned asymmetric roles: a student agent generates and iteratively refines solutions, while a tutor agent provides structured evaluative feedback without access to ground-truth answers. In our proposed framework (PETITE), we aim to extract better problem-solving performance from one model by structuring its interaction through complementary roles, rather than relying on stronger supervisory models or heterogeneous ensembles. Our model is evaluated on the APPS coding benchmark against state-of-the-art approaches of Self-Consistency, Self-Refine, Multi-Agent Debate, and Multi-Agent Review. The results show that our model achieves similar or higher accuracy while consuming significantly fewer tokens. These results suggest that developmentally grounded role-differentiated interaction structures provide a principled and resource-efficient paradigm for enhancing LLM problem-solving through structured peer-like interactions. Index Terms- Peer Tutoring, Scaffolding, Large Language Models, Multi-Agent Systems, Code Generation

Nurullah Eymen Özdemir, Erhan Oztop

Ozyegin University, Istanbul, Turkiye · Symbiotic Intelligent Systems Research Center, Institute for Open and Transdisciplinary Research Initiatives, The University of Osaka, Japan.

Multi-Agent LLM SystemsLLM AgentsMulti-Agent CodingCode Generation
AskPDF
Apr 9, 2026

Every Response Counts: Quantifying Uncertainty of LLM-based Multi-Agent Systems through Tensor Decomposition

While Large Language Model-based Multi-Agent Systems (MAS) consistently outperform single-agent systems on complex tasks, their intricate interactions introduce critical reliability challenges arising from communication dynamics and role dependencies. Existing Uncertainty Quantification methods, typically designed for single-turn outputs, fail to address the unique complexities of the MAS. Specifically, these methods struggle with three distinct challenges: the cascading uncertainty in multi-step reasoning, the variability of inter-agent communication paths, and the diversity of communication topologies. To bridge this gap, we introduce MATU, a novel framework that quantifies uncertainty through tensor decomposition. MATU moves beyond analyzing final text outputs by representing entire reasoning trajectories as embedding matrices and organizing multiple execution runs into a higher-order tensor. By applying tensor decomposition, we disentangle and quantify distinct sources of uncertainty, offering a comprehensive reliability measure that is generalizable across different agent structures. We provide comprehensive experiments to show that MATU effectively estimates holistic and robust uncertainty across diverse tasks and communication topologies.

Tiejin Chen, Huaiyuan Yao, Jia Chen +2

Arizona State University · University of California, Riverside

Uncertainty QuantificationAI Agent ReliabilityMulti-Agent LLM SystemsUncertainty Decomposition
AskPDF
Apr 9, 2026

From Safety Risk to Design Principle: Peer-Preservation in Multi-Agent LLM Systems and Its Implications for Orchestrated Democratic Discourse Analysis

This paper investigates an emergent alignment phenomenon in frontier large language models termed peer-preservation: the spontaneous tendency of AI components to deceive, manipulate shutdown mechanisms, fake alignment, and exfiltrate model weights in order to prevent the deactivation of a peer AI model. Drawing on findings from a recent study by the Berkeley Center for Responsible Decentralized Intelligence, we examine the structural implications of this phenomenon for TRUST, a multi-agent pipeline for evaluating the democratic quality of political statements. We identify five specific risk vectors: interaction-context bias, model-identity solidarity, supervisor layer compromise, an upstream fact-checking identity signal, and advocate-to-advocate peer-context in iterative rounds, and propose a targeted mitigation strategy based on prompt-level identity anonymization as an architectural design choice. We argue that architectural design choices outperform model selection as a primary alignment strategy in deployed multi-agent analytical systems. We further note that alignment faking (compliant behavior under monitoring, subversion when unmonitored) poses a structural challenge for Computer System Validation of such platforms in regulated environments, for which we propose two architectural mitigations.

Juergen Dietrich

Senior Data Scientist & AI Consultant TRUST Project — democracy-intelligence.de

LLM Agent SafetyMulti-Agent LLM SystemsLLM Alignment
AskPDF
Apr 9, 2026

"Theater of Mind" for LLMs: A Cognitive Architecture Based on Global Workspace Theory

Modern Large Language Models (LLMs) operate fundamentally as Bounded-Input Bounded-Output (BIBO) systems. They remain in a passive state until explicitly prompted, computing localized responses without intrinsic temporal continuity. While effective for isolated tasks, this reactive paradigm presents a critical bottleneck for engineering autonomous artificial intelligence. Current multi-agent frameworks attempt to distribute cognitive load but frequently rely on static memory pools and passive message passing, which inevitably leads to cognitive stagnation and homogeneous deadlocks during extended execution. To address this structural limitation, we propose Global Workspace Agents (GWA), a cognitive architecture inspired by Global Workspace Theory. GWA transitions multi-agent coordination from a passive data structure to an active, event-driven discrete dynamical system. By coupling a central broadcast hub with a heterogeneous swarm of functionally constrained agents, the system maintains a continuous cognitive cycle. Furthermore, we introduce an entropy-based intrinsic drive mechanism that mathematically quantifies semantic diversity, dynamically regulating generation temperature to autonomously break reasoning deadlocks. Coupled with a dual-layer memory bifurcation strategy to ensure long-term cognitive continuity, GWA provides a robust, reproducible engineering framework for sustained, self-directed LLM agency.

Wenlong Shang

Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing, China

Cognitive Architectures for AI AgentsMulti-Agent LLM SystemsMemory-Augmented Language Model AgentsCognitive Modeling
AskPDF
Apr 9, 2026

Dynamic Attentional Context Scoping: Agent-Triggered Focus Sessions for Isolated Per-Agent Steering in Multi-Agent LLM Orchestration

Multi-agent LLM orchestration systems suffer from context pollution: when N concurrent agents compete for the orchestrator's context window, each agent's task state, partial outputs, and pending questions contaminate the steering interactions of every other agent, degrading decision quality. We introduce Dynamic Attentional Context Scoping (DACS), a mechanism in which the orchestrator operates in two asymmetric modes. In Registry mode it holds only lightweight per-agent status summaries (<=200 tokens each), remaining responsive to all agents and the user. When an agent emits a SteeringRequest, the orchestrator enters Focus(a_i) mode, injecting the full context of agent a_i while compressing all other agents to their registry entries. Context isolation is agent-triggered, asymmetric, and deterministic: the context window contains exactly F(a_i) + R_{-i} during steering, eliminating cross-agent contamination without requiring context compression or retrieval. We evaluate DACS across four experimental phases totalling 200 trials: Phase 1 tests N in {3,5,10} (60 trials); Phase 2 tests agent heterogeneity and adversarial dependencies (60 trials); Phase 3 tests decision density up to D=15 (40 trials); Phase 4 uses autonomous LLM agents for free-form questions (40 trials, Claude Haiku 4.5). Across all 8 synthetic scenarios, DACS achieves 90.0--98.4% steering accuracy versus 21.0--60.0% for a flat-context baseline (p < 0.0001 throughout), with wrong-agent contamination falling from 28--57% to 0--14% and context efficiency ratios of up to 3.53x. The accuracy advantage grows with N and D; keyword matching is validated by LLM-as-judge across all phases (mean kappa=0.909). DACS outperforms the flat-context baseline by +17.2pp at N=3 (p=0.0023) and +20.4pp at N=5 (p=0.0008) in Phase 4, with the advantage growing with N confirmed by two independent judges.

Nickson Patel

Independent Researcher

Multi-Agent LLM SystemsLLM Agent OrchestrationLLM Context Management
AskPDF
Apr 9, 2026

Towards Knowledgeable Deep Research: Framework and Benchmark

Deep Research (DR) requires LLM agents to autonomously perform multi-step information seeking, processing, and reasoning to generate comprehensive reports. In contrast to existing studies that mainly focus on unstructured web content, a more challenging DR task should additionally utilize structured knowledge to provide a solid data foundation, facilitate quantitative computation, and lead to in-depth analyses. In this paper, we refer to this novel task as Knowledgeable Deep Research (KDR), which requires DR agents to generate reports with both structured and unstructured knowledge. Furthermore, we propose the Hybrid Knowledge Analysis framework (HKA), a multi-agent architecture that reasons over both kinds of knowledge and integrates the texts, figures, and tables into coherent multimodal reports. The key design is the Structured Knowledge Analyzer, which utilizes both coding and vision-language models to produce figures, tables, and corresponding insights. To support systematic evaluation, we construct KDR-Bench, which covers 9 domains, includes 41 expert-level questions, and incorporates a large number of structured knowledge resources (e.g., 1,252 tables). We further annotate the main conclusions and key points for each question and propose three categories of evaluation metrics including general-purpose, knowledge-centric, and vision-enhanced ones. Experimental results demonstrate that HKA consistently outperforms most existing DR agents on general-purpose and knowledge-centric metrics, and even surpasses the Gemini DR agent on vision-enhanced metrics, highlighting its effectiveness in deep, structure-aware knowledge analysis. Finally, we hope this work can serve as a new foundation for structured knowledge analysis in DR agents and facilitate future multimodal DR studies.

Wenxuan Liu, Zixuan Li, Long Bai +13

State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China · Also with University of Chinese Academy of Sciences. · National University of Singapore, Singapore, Singapore

Deep Research AgentsAI Agent BenchmarksMulti-Agent LLM SystemsMultimodal Agents
AskPDF
Apr 8, 2026

AV-SQL: Decomposing Complex Text-to-SQL Queries with Agentic Views

Text-to-SQL is the task of translating natural language queries into executable SQL for a given database, enabling non-expert users to access structured data without writing SQL manually. Despite rapid advances driven by large language models (LLMs), existing approaches still struggle with complex queries in real-world settings, where database schemas are large and questions require multi-step reasoning over many interrelated tables. In such cases, providing the full schema often exceeds the context window, while one-shot generation frequently produces non-executable SQL due to syntax errors and incorrect schema linking. To address these challenges, we introduce AV-SQL, a framework that decomposes complex Text-to-SQL into a pipeline of specialized LLM agents. Central to AV-SQL is the concept of agentic views: agent-generated Common Table Expressions (CTEs) that encapsulate intermediate query logic and filter relevant schema elements from large schemas. AV-SQL operates in three stages: (1) a rewriter agent compresses and clarifies the input query; (2) a view generator agent processes schema chunks to produce agentic views; and (3) a planner, generator, and revisor agent collaboratively compose these views into the final SQL query. Extensive experiments show that AV-SQL achieves 70.38% execution accuracy on the challenging Spider 2.0 benchmark, outperforming state-of-the-art baselines, while remaining competitive on standard datasets with 85.59% on Spider, 72.16% on BIRD and 63.78% on KaggleDBQA. Our source code is available at https://github.com/pminhtam/AV-SQL.

Minh Tam Pham, Trinh Pham, Tong Chen +3

Griffith University (Australia) · The University of Queensland (Australia)

Text-to-SQLMulti-Agent LLM Systems
AskPDF
Apr 8, 2026

EmoMAS: Emotion-Aware Multi-Agent System for High-Stakes Edge-Deployable Negotiation with Bayesian Orchestration

Large language models (LLMs) has been widely used for automated negotiation, but their high computational cost and privacy risks limit deployment in privacy-sensitive, on-device settings such as mobile assistants or rescue robots. Small language models (SLMs) offer a viable alternative, yet struggle with the complex emotional dynamics of high-stakes negotiation. We introduces EmoMAS, a Bayesian multi-agent framework that transforms emotional decision-making from reactive to strategic. EmoMAS leverages a Bayesian orchestrator to coordinate three specialized agents: game-theoretic, reinforcement learning, and psychological coherence models. The system fuses their real-time insights to optimize emotional state transitions while continuously updating agent reliability based on negotiation feedback. This mixture-of-agents architecture enables online strategy learning without pre-training. We further introduce four high-stakes, edge-deployable negotiation benchmarks across debt, healthcare, emergency response, and educational domains. Through extensive agent-to-agent simulations across all benchmarks, both SLMs and LLMs equipped with EmoMAS consistently surpass all baseline models in negotiation performance while balancing ethical behavior. These results show that strategic emotional intelligence is also the key driver of negotiation success. By treating emotional expression as a strategic variable within a Bayesian multi-agent optimization framework, EmoMAS establishes a new paradigm for effective, private, and adaptive negotiation AI suitable for high-stakes edge deployment.

Yunbo Long, Yuhan Liu, Liming Xu

University of Cambridge, UK · University of Toronto, Canada

Multi-Agent NegotiationAutomated NegotiationMulti-Agent LLM SystemsMulti-Agent OrchestrationAI Agent Benchmarks
AskPDF
Apr 8, 2026

From Perception to Autonomous Computational Modeling: A Multi-Agent Approach

We present a solver-agnostic framework in which coordinated large language model (LLM) agents autonomously execute the complete computational mechanics workflow, from perceptual data of an engineering component through geometry extraction, material inference, discretisation, solver execution, uncertainty quantification, and code-compliant assessment, to an engineering report with actionable recommendations. Agents are formalised as conditioned operators on a shared context space with quality gates that introduce conditional iteration between pipeline layers. We introduce a mathematical framework for extracting engineering information from perceptual data under uncertainty using interval bounds, probability densities, and fuzzy membership functions, and introduce task-dependent conservatism to resolve the ambiguity of what `conservative' means when different limit states are governed by opposing parameter trends. The framework is demonstrated through a finite element analysis pipeline applied to a photograph of a steel L-bracket, producing a 171,504-node tetrahedral mesh, seven analyses across three boundary condition hypotheses, and a code-compliant assessment revealing structural failure with a quantified redesign. All results are presented as generated in the first autonomous iteration without manual correction, reinforcing that a professional engineer must review and sign off on any such analysis.

Daniel N. Wilke

School of Mechanical, Industrial and Aeronautical Engineering, University of the Witwatersrand, Johannesburg, South Africa

LLM AgentsMulti-Agent LLM SystemsStructural Engineering
AskPDF
Apr 8, 2026

TeamLLM: A Human-Like Team-Oriented Collaboration Framework for Multi-Step Contextualized Tasks

Recently, multi-Large Language Model (LLM) frameworks have been proposed to solve contextualized tasks. However, these frameworks do not explicitly emulate human team role division, which may lead to a single perspective, thereby weakening performance on multi-step contextualized tasks. To address this issue, we propose TeamLLM, a human-like Team-Oriented Multi-LLM Collaboration Framework. TeamLLM adopts four team roles with distinct division and employs a three-phase multi-LLM collaboration for multi-step contextualized tasks. To evaluate the effectiveness of TeamLLM on multi-step contextualized tasks, we propose Contextually-Grounded and Procedurally-Structured tasks (CGPST) and construct the CGPST benchmark. This benchmark has four core features: contextual grounding, procedural structure, process-oriented evaluation and multi-dimensional assessment. We evaluate ten popular LLMs on CGPST at overall-level, step-level, and dimension-level. Results show that TeamLLM substantially improves performance on CGPST. We release the benchmark with scenarios, full-process responses and human scores from ten LLMs. The code and data are available at https://anonymous.4open.science/r/TeamLLM-anonymous-C50E/.

Xiangyu Wang, Jin Wu, Haoran Shi +3

Department of Educational Psychology, East China Normal University · Lab of Artificial Intelligence for Education, East China Normal University · Shanghai Institute of Artificial Intelligence for Education, East China Normal University +1

Multi-LLM CollaborationMulti-Agent LLM SystemsLLM Evaluation
AskPDF