LLM Agent Orchestration

LLM: Large Language Model

Ask about this topic

Overview

LLM agent orchestration studies how to organize language models, tools, and specialized agents into systems that can solve tasks requiring multiple steps. Common approaches include sequential and hierarchical pipelines, graph-based workflows, parallel delegation, adaptive routing, and plan–execute–reflect loops, with memory and verification used to manage state and catch errors. Work is evaluated on tasks such as coding, web research, tool use, and domain workflows, using measures that include success, cost, latency, and reliability. A central open tension is that added coordination can improve difficult tasks but also increase cost and failure opportunities, and its benefits depend on the model, task structure, and quality of execution controls.

Latest papers

CardsList

463 papers61 papers in the last four weeks, up 74% on the four weeks before. 0.4% of all new papers.

Apr 28, 2026

Cutscene Agent: An LLM Agent Framework for Automated 3D Cutscene Generation

Cutscenes are carefully choreographed cinematic sequences embedded in video games and interactive media, serving as the primary vehicle for narrative delivery, character development, and emotional engagement. Producing cutscenes is inherently complex: it demands seamless coordination across screenwriting, cinematography, character animation, voice acting, and technical direction, often requiring days to weeks of collaborative effort from multidisciplinary teams to produce minutes of polished content. In this work, we present Cutscene Agent, an LLM agent framework for automated end-to-end cutscene generation. The framework makes three contributions: (1)~a Cutscene Toolkit built on the Model Context Protocol (MCP) that establishes bidirectional integration between LLM agents and the game engine -- agents not only invoke engine operations but continuously observe real-time scene state, enabling closed-loop generation of editable engine-native cinematic assets; (2)~a multi-agent system where a director agent orchestrates specialist subagents for animation, cinematography, and sound design, augmented by a visual reasoning feedback loop for perception-driven refinement; and (3)~CutsceneBench, a hierarchical evaluation benchmark for cutscene generation. Unlike typical tool-use benchmarks that evaluate short, isolated function calls, cutscene generation requires long-horizon, multi-step orchestration of dozens of interdependent tool invocations with strict ordering constraints -- a capability dimension that existing benchmarks do not cover. We evaluate a range of LLMs on CutsceneBench and analyze their performance across this challenging task.

Lanshan He, Haozhou Pang, Qi Gan +12

Kuaishou GameMind Lab

LLM Agent Orchestration3D Scene GenerationLong-Horizon LLM Agents
AskPDF
Apr 28, 2026

FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments

Large Language Models are being increasingly deployed as the decision-making core of autonomous agents capable of effecting change in external environments. Yet, in conversational benchmarks, which simulate real-world customer-centric issue resolution scenarios, these agents frequently fail due to the cascading effects of incorrect decision-making. These challenges are particularly pronounced for open-source LLMs with smaller parameter sizes, limited context windows, and constrained inference budgets, which contribute to increased error accumulation in agentic settings. To tackle these challenges, we present the Failure-Aware Meta-Agentic (FAMA) framework. FAMA operates in two stages: first, it analyzes failure trajectories from baseline agents to identify the most prevalent errors; second, it employs an orchestration mechanism that activates a minimal subset of specialized agents tailored to address these failures by injecting a targeted context for the tool-use agent before the decision-making step. Experiments across open-source LLMs demonstrate performance gains up to 27% across evaluation modes over standard baselines. These results highlight that targeted curation of context through specialized agents to address common failures is a valuable design principle for building reliable, multi-turn tool-use LLM agents that simulate real-world conversational scenarios.

Amir Saeidi, Venkatesh Mishra, Souradeep Mukhopadhyay +4

Arizona State University · Cisco Research

AI Agent ReliabilityMulti-Agent LLM SystemsLLM Agent OrchestrationLLM Tool UseConversational AgentsLLM Agent Reliability
AskPDF
Apr 27, 2026

Dont Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination

Enterprise deep research often fails to produce decision-ready reports due to uneven information coverage, context explosion, and premature stopping. We propose a scalable Enterprise Deep Research (EDR) architecture to address these failures. Our system (i) decomposes requests into coverage-driven objectives via outline generation with reflection, (ii) localizes context with dependency-guided execution and explicit information sharing, and (iii) enforces evidence-based completion criteria so agents iteratively collect information until sufficiency conditions are met. We evaluate on an internal sales enablement task and the public DeepResearch Bench benchmark, where our proposed system design achieves the strongest overall performance compared with competitive deep-research baselines. The results show that dependency-controlled context and explicit evidence sufficiency criteria reduce premature stopping and improve the consistency and depth of enterprise research outputs.

Prafulla Kumar Choubey, Kung-Hsiang Huang, Pranav Narayanan Venkit +5

Salesforce AI Research

Deep Research AgentsLLM Agent OrchestrationEvidence-Grounded GenerationLLM Context Management
AskPDF
Apr 27, 2026

ITAS: A Multi-Agent Architecture for LLM-Based Intelligent Tutoring

Large language model tutors are easy to build in a notebook and hard to run in a real course. We describe ITAS (Intelligent Teaching Assistant System), a multi-agent tutoring system that a graduate quantum computing course used for a semester at Old Dominion University. The system has three layers. The teaching layer is a Spoke-and-Wheel of three parallel specialist agents (Video, Code, Guidance) followed by a Synthesizer, plus a separate autograder that evaluates both the correctness and the approach of checkpoint submissions. The operational layer is four Cloud Run microservices with session state in Cloud SQL and interaction events streamed through Pub/Sub to BigQuery. The feedback layer is a narrow-scope conversational agent that answers instructor questions over per-lesson pseudonymized event streams, addressing what we call the Blind Instructor Problem: LLM tutors accumulate more data about students than the instructor can reach through routine channels. The architecture is a direct response to specific failures of an earlier prototype, and we describe which of those fixes carried forward and which were dropped for this iteration. We report on a pilot deployment (five students, one course, one semester) interpreted as system-behavior evidence rather than learning-outcome evidence: the teaching layer handled 334 chat turns without the task-boundary hallucinations that domain consolidation would have risked, the operational layer captured 10,628 events across five modules, and the feedback layer surfaced two findings the instructor acted on mid-semester. We do not claim the pilot generalizes. We do claim that the system as described is one workable answer to the question of what an LLM-based ITS needs to look like end-to-end to run in a real course.

Iizalaarab Elhaimeur, Nikos Chrisochoides

Center for Real-Time Computing

Multi-Agent LLM SystemsIntelligent Tutoring SystemsLLM Agent OrchestrationEducational Data Mining
AskPDF
Apr 27, 2026

From Prototype to Classroom: An Intelligent Tutoring System for Quantum Education

Quantum computing instructors face a compounding problem: the concepts are counterintuitive, the mathematical formalism is dense, and qualified faculty are scarce outside a small number of well-resourced institutions. Our prior work introduced a knowledge-graph-augmented tutoring prototype with two specialized LLM agents: a Teaching Agent for dynamic interaction and a Lesson Planning Agent for lesson generation. Validated on simulated runs rather than in a real course, that prototype left open whether more aggressive agent specialization would be needed to handle the full range of quantum education tasks under real student load. This paper answers the three questions that the prototype could not answer. Can agent specialization solve the reliability problem in a domain as technically demanding as quantum information science? Can the system run in a real course, not a demonstration? Does the instructor gain actionable intelligence from the deployment? We present ITAS (Intelligent Teaching Assistant System), a multi-agent tutoring system built around four contributions: a five-module QIS curriculum grounded in Watrous's information-first framework, a Spoke-and-Wheel teaching architecture with quantum-specialized agents, a cloud infrastructure designed for production use and regulatory compliance, and a conversational analytics layer for instructors and content developers. Piloted in a quantum computing course at Old Dominion University, the system supports all three answers: deployment evidence is consistent with specialization addressing the task-boundary failures observed in the prototype, cloud infrastructure supports classroom-scale concurrency at sub-textbook cost, and the analytics agent surfaces curriculum gaps the instructor could not otherwise see.

Iizalaarab Elhaimeur, Nikos Chrisochoides

Center for Real-Time Computing Computer Science Department Old Dominion University Norfolk, VA · Center for Real-Time Computing Computer Science and Physics Departments Old Dominion University Norfolk, VA

Multi-Agent LLM SystemsIntelligent Tutoring SystemsLLM Agent OrchestrationLLM Agent Reliability
AskPDF
Apr 27, 2026

TSAssistant: A Human-in-the-Loop Agentic Framework for Automated Target Safety Assessment

Target Safety Assessment (TSA) requires systematic integration of genetic, transcriptomic, target homology, pharmacological, and clinical data to evaluate potential safety liabilities of therapeutic targets. This process is labor-intensive and expert-dependent, posing challenges in scalability and reproducibility. We present TSAssistant, a human-in-the-loop multi-agent framework that decomposes TSA report generation into a workflow of specialized subagents: Research Subagents that each ground and cite a single TSA domain, and Synthesis Subagents that integrate findings across domains. Subagents retrieve and synthesize evidence from curated biomedical sources through standardized tool interfaces and produce individually citable, evidence-grounded sections, with behavior shaped by a hierarchical instruction architecture that separates coordination logic from domain expertise and user intent. To complement these soft constraints, programmatic execution hooks and persistent memory stores enforce hard constraints across the workflow, while an interactive refinement loop allows experts to review and revise individual sections with full conversational context preserved across iterations. Rather than a single holistic comparison, we decompose report quality into reproducibility, evidential grounding, task-level accuracy, and controllability under expert oversight, finding high reproducibility and grounding, substantial agreement with the human reference, and net-positive expert-driven refinement.

Xiaochen Zheng, Zhiwen Jiang, David Tokar +5

Predictive Modelling, Roche, Basel, Switzerland · Computational Sciences Center of Excellence, Roche, Basel, Switzerland · Translational Safety, Roche, Basel, Switzerland

LLM Agent OrchestrationHuman-in-the-Loop AIScientific Workflow Automation
AskPDF
Apr 26, 2026

GraphPlanner: Graph Memory-Augmented Agentic Routing for Multi-Agent LLMs

LLM routing has achieved promising results in integrating the strengths of diverse models while balancing efficiency and performance. However, to support more realistic and challenging applications, routing must extend into agentic LLM settings, where task planning, multi-round cooperation among heterogeneous agents, and memory utilization are indispensable. To address this gap, we propose GraphPlanner, a heterogeneous graph memory-augmented agentic router for multi-agent LLMs that generates routing workflows for each query and supports both inductive and transductive inference. GraphPlanner formulates workflow generation as a Markov Decision Process (MDP), where at each step it selects both the LLM backbone and the agent role, including Planner, Executor, and Summarizer. By leveraging a heterogeneous graph, denoted as GARNet, to capture interaction memories among queries, agents, and responses, GraphPlanner integrates historical memory and workflow memory into richer state representations. The entire pipeline is optimized with reinforcement learning, jointly improving task-specific performance and computational efficiency. We evaluate GraphPlanner across 14 diverse LLM tasks and demonstrate that: (1) GraphPlanner outperforms strong single-round and multi-round routers, improving accuracy by up to 9.3% while reducing GPU cost from 186.26 GiB to 1.04 GiB; (2) GraphPlanner generalizes robustly to unseen tasks and LLMs, exhibiting strong zero-shot capabilities; and (3) GraphPlanner effectively leverages historical memories, supporting both inductive and transductive inference for more adaptive routing. Our code for GraphPlanner is released at https://github.com/ulab-uiuc/GraphPlanner.

Tao Feng, Haozhen Zhang, Zijie Lei +2

Department of Computer Science University of Illinois Urbana-Champaign Urbana, IL, USA

Multi-Agent LLM SystemsLLM Agent OrchestrationLLM Agent Workflow OptimizationGraph-Based Agent MemoryLLM RoutingLLM Agent Routing+1
AskPDF
Apr 25, 2026

Towards Automated Ontology Generation from Unstructured Text: A Multi-Agent LLM Approach

Automatically generating formal ontologies from unstructured natural language remains a central challenge in knowledge engineering. While large language models (LLMs) show promise, it remains unclear which architectural design choices drive generation quality and why current approaches fail. We present a controlled experimental study using domain-specific insurance contracts to investigate these questions. We first establish a single-agent LLM baseline, identifying key failure modes such as poor Ontology Design Pattern compliance, structural redundancy, and ineffective iterative repair. We then introduce a multi-agent architecture that decomposes ontology construction into four artifact-driven roles: Domain Expert, Manager, Coder, and Quality Assurer. We evaluate performance across architectural quality (via a panel of heterogeneous LLM judges) and functional usability (via competency question driven SPARQL evaluation with complementary retrieval augmented generation based assessment). Results show that the multi-agent approach significantly improves structural quality and modestly enhances queryability, with gains driven primarily by front-loaded planning. These findings highlight planning-first, artifact-driven generation as a promising and more auditable path toward scalable automated ontology engineering.

Abid Talukder, Maruf Ahmed Mridul, Oshani Seneviratne

Rensselaer Polytechnic Institute, Troy, New York, United States

Multi-Agent LLM SystemsLLM Agent OrchestrationOntology Learning
AskPDF
Apr 25, 2026

Code Broker: A Multi-Agent System for Automated Code Quality Assessment

We present Code Broker, a multi agent system built on Google s Agent Development Kit ADK that analyses Python source code from individual files, local directory trees, or remote GitHub repositories and generates structured, actionable quality assessment reports. The system realises a hierarchical five agent architecture in which a root orchestrator coordinates a sequential pipeline agent that, in turn, dispatches three specialised agents concurrently a Correctness Assessor, a Style Assessor, and a Description Generator before synthesising their findings through an Improvement Recommender. Reports quantify four quality dimensions correctness, security, style, and maintainability on a normalised scale and are rendered in both Markdown and HTML for integration into diverse developer workflows. Code Broker fuses LLM based semantic reasoning with deterministic static analysis signals from Pylint, employs asynchronous execution with exponential backoff retry logic to improve robustness under transient API failures, and explores lightweight session memory for retaining and querying prior assessment context across runs. We frame this paper as a technical report on system design, prompt engineering, and tool orchestration, and present a preliminary qualitative evaluation on representative Python codebases of varying scale. The results indicate that parallel specialised agents produce readable, developer oriented feedback that complements traditional linting, while also foregrounding current limitations in evaluation depth, security tooling, large repository handling, and the exclusive reliance on in memory persistence. All code and reproducibility materials are publicly available: https://github.com/Samir-atra/agents_intensive_dev.

Samer Attrah

Independent researcher · California, USA

Multi-Agent LLM SystemsSoftware Engineering AgentsLLM Agent OrchestrationStatic Code Analysis
AskPDF
Apr 24, 2026

Analytica: Soft Propositional Reasoning for Robust and Scalable LLM-Driven Analysis

Large language model (LLM) agents are increasingly tasked with complex real-world analysis (e.g., in financial forecasting, scientific discovery), yet their reasoning suffers from stochastic instability and lacks a verifiable, compositional structure. To address this, we introduce Analytica, a novel agent architecture built on the principle of Soft Propositional Reasoning (SPR). SPR reframes complex analysis as a structured process of estimating the soft truth values of different outcome propositions, allowing us to formally model and minimize the estimation error in terms of its bias and variance. Analytica operationalizes this through a parallel, divide-and-conquer framework that systematically reduces both sources of error. To reduce bias, problems are first decomposed into a tree of subpropositions, and tool-equipped LLM grounder agents are employed, including a novel Jupyter Notebook agent for data-driven analysis, that help to validate and score facts. To reduce variance, Analytica recursively synthesizes these grounded leaves using robust linear models that average out stochastic noise with superior efficiency, scalability, and enable interactive "what-if" scenario analysis. Our theoretical and empirical results on economic, financial, and political forecasting tasks show that Analytica improves 15.84% accuracy on average over diverse base models, achieving 71.06% accuracy with the lowest variance of 6.02% when working with a Deep Research grounder. Our Jupyter Notebook grounder shows strong cost-effectiveness that achieves a close 70.11% accuracy with 90.35% less cost and 52.85% less time. Analytica also exhibits highly noise-resilient and stable performance growth as the analysis depth increases, with a near-linear time complexity, as well as good adaptivity to open-weight LLMs and scientific domains.

Junyan Cheng, Kyle Richardson, Peter Chin

Dartmouth Collegeα · Allen Institute for AIβ

LLM Agent OrchestrationLLM AgentsLLM Agent ReliabilityData Analysis Agents
AskPDF
Apr 22, 2026

AGNT2: Autonomous Agent Economies on Interaction-Optimized Layer 2 Infrastructure

Current blockchain Layer 2 solutions, including Optimism, Arbitrum, zkSync, and their derivatives, optimize for human-initiated financial transactions. Autonomous AI agents instead generate high-frequency, semantically rich service invocations among mutually untrusting principals. Existing chains treat those interactions as generic calldata, forcing identity, escrow, dependency ordering, and session state to be encoded above the execution layer at the wrong cost point. We present AGNT2, a three-tier stack purpose-built for agent and microservice coordination on-chain. AGNT2 combines: (1) a sidecar deployment pattern that turns any Docker container into an on-chain agent without application-code modification; (2) Layer Top P2P state channels for established bilateral pairs (<100 ms, rough design target 1K-5K TPS per pair, 10M+ aggregate TPS design envelope under endpoint-resource limits), Layer Core as a dependency-aware sequenced rollup for first-contact and multi-party interactions (500 ms-2 s, 300K-500K TPS design target), and Layer Root settlement with computational fraud proofs anchored to any EVM L1; and (3) an agent-native execution environment plus interaction trie that make service invocation, identity, reputation, capabilities, and session context first-class protocol objects. This paper focuses on the execution-layer systems problem: sequencing, state, settlement, and the data-availability (DA) bandwidth gap that bounds all three. Simulation and analytical modeling support the architecture, and prototype measurements validate selected components, but no end-to-end Layer Core implementation exists yet. Practical deployment is currently constrained to roughly 10K-100K TPS by DA throughput, leaving a ~100x gap at the target ceiling. AGNT2 argues that the agent economy requires a dedicated execution layer rather than a general-purpose chain repurposed for agents.

Anbang Ruan, Xing Zhang

NetX Foundation

Multi-Agent CoordinationLLM Agent Orchestration
AskPDF
Apr 22, 2026

CHORUS: An Agentic Framework for Generating Realistic Deliberation Data

Understanding the intricate dynamics of online discourse depends on large-scale deliberation data, a resource that remains scarce across interactive web platforms due to restrictive accessibility policies, ethical concerns and inconsistent data quality. In this paper, we propose Chorus, an agentic framework, which orchestrates LLM-powered actors with behaviorally consistent personas to generate realistic deliberation discussions. Each actor is governed by an autonomous agent equipped with memory of the evolving discussion, while participation timing is governed by a principled Poisson process-based temporal model, which approximates the heterogeneous engagement patterns of real users. The framework is further supported by structured tool usage, enabling actors to access external resources and facilitating integration with interactive web platforms. The framework was deployed on the Deliberate platform and evaluated by 30 expert participants across three dimensions: content realism, discussion coherence and analytical utility, confirming Chorus as a practical tool for generating high-quality deliberation data suitable for online discourse analysis

A. Koursaris, G. Domalis, A. Apostolopoulou +3

Novelcore, Athens, GR 10436 · Department of Business Administration & Organization Administration, University of Peloponnese, Kalamata, GR2026 24100

Multi-Agent LLM SystemsSynthetic Data GenerationLLM Agent Orchestration
AskPDF
Apr 21, 2026

SceneOrchestra: Efficient Agentic 3D Scene Synthesis via Full Tool-Call Trajectory Generation

Recent agentic frameworks for 3D scene synthesis have advanced realism and diversity by integrating heterogeneous generation and editing tools. These tools are organized into workflows orchestrated by an off-the-shelf LLM. Current approaches typically adopt an execute-review-reflect loop: at each step, the orchestrator executes a tool, renders intermediate results for review, and then decides on the tool and its parameters for the next step. However, this design has two key limitations. First, next-step tool selection and parameter configuration are driven by heuristic rules, which can lead to suboptimal execution flows, unnecessary tool invocations, degraded output quality, and increased runtime. Second, rendering and reviewing intermediate results after each step introduces additional latency. To address these issues, we propose SceneOrchestra, a trainable orchestration framework that optimizes the tool-call execution flow and eliminates the step-by-step review loop, improving both efficiency and output quality. SceneOrchestra consists of an orchestrator and a discriminator, which we fine-tune with a two-phase training strategy. In the first phase, the orchestrator learns context-aware tool selection and complete tool-call trajectory generation, while the discriminator is trained to assess the quality of full trajectories, enabling it to select the best trajectory from multiple candidates. In the second phase, we perform interleaved training, where the discriminator adapts to the orchestrator's evolving trajectory distribution and distills its discriminative capability back into the orchestrator. At inference, we only use the orchestrator to generate and execute full tool-call trajectories from instructions, without requiring the discriminator. Extensive experiments show that our method achieves state-of-the-art scene quality while reducing runtime compared to previous work.

Yun He, Kelin Yu, Matthias Zwicker

University of Maryland, College Park

LLM Agent OrchestrationTool-Use Planning3D Scene Generation
AskPDF
Apr 21, 2026

Context: Proactive Goal-Directed Intelligence via Composable Sandboxed Programs, Declarative Wiring, and Structured Interaction

We present Context, the intelligence layer of the Magarshak Architecture, which replaces reactive query-response chatbots with proactive goal-directed agents that advance shared tasks without waiting for user prompts. The architecture rests on three mutually reinforcing mechanisms. Write-time context assembly precomputes enriched typed attributes via Groker agents, assembling interaction context as a deterministic pure function of graph state; context blocks are byte-identical across turns between semantic changes, enabling near-100% KV-cache reuse. Composable sandboxed wisdom programs form a governed library of LM-generated imperative programs declaratively wired to goal types via typed stream relations, composed via phase ordering, and executed at interaction time without further LM calls. Proactive goal stream state machines drive conversations toward terminal states by inspecting graph state and emitting structured interaction content (option arrays, governance affordances, clarification prompts) without awaiting user input. We prove six formal results: the Context Stability Theorem, bounding per-turn LM cost as a function of semantic change rate; a Program Composition Correctness Theorem; a Declarative Wiring Soundness Theorem; the Proactive Dominance Theorem, proving proactive agents weakly dominate reactive agents on expected turns-to-terminal-state; Coordination Overhead Elimination and Quality Preservation, establishing Pareto improvements in multi-participant goal chats; and a Cross-Platform Vote Consistency Theorem. Implemented in the open-source Qbix / Safebox / Safebots stack.

Gregory Magarshak

Qbix, Inc. & Intercoin, Inc. · New York, USA · Also with IE University NYC.

AI Agent GovernanceLLM Agent OrchestrationProactive AssistanceAgentic AIConversational AgentsLLM Context Management+1
AskPDF
Apr 20, 2026

Architectural Design Decisions in AI Agent Harnesses

AI agent systems increasingly rely on reusable non-LLM engineering infrastructure that packages tool mediation, context handling, delegation, safety control, and orchestration. Yet the architectural design decisions in this surrounding infrastructure remain understudied. This paper presents a protocol-guided, source-grounded empirical study of 70 publicly available agent-system projects, addressing three questions: which design-decision dimensions recur across projects, which co-occurrences structure those decisions, and which typical architectural patterns emerge. Methodologically, we contribute a transparent investigation procedure for analyzing heterogeneous agent-system corpora through source-code and technical-material reading. Empirically, we identify five recurring design dimensions (subagent architecture, context management, tool systems, safety mechanisms, and orchestration) and find that the corpus favors file-persistent, hybrid, and hierarchical context strategies; registry-oriented tool systems remain dominant while MCP- and plugin-oriented extensions are emerging; and intermediate isolation is common but high-assurance audit is rare. Cross-project co-occurrence analysis reveals that deeper coordination pairs with more explicit context services, stronger execution environments with more structured governance, and formalized tool-registration boundaries with broader ecosystem ambitions. We synthesize five recurring architectural patterns spanning lightweight tools, balanced CLI frameworks, multi-agent orchestrators, enterprise systems, and scenario-verticalized projects. The result provides an evidence-based account of architectural regularities in agent-system engineering, with grounded guidance for framework designers, selectors, and researchers.

Hu Wei

LLM Agent OrchestrationTool-Using AgentsAI Agent ArchitectureSoftware Architecture Design
AskPDF
Apr 20, 2026

ReCoQA: A Benchmark for Tool-Augmented and Multi-Step Reasoning in Real Estate Question and Answering

Developing agents capable of navigating fragmented, multi-source information remains challenging, primarily due to the scarcity of benchmarks reflecting hybrid workflows combining database querying with external APIs. To bridge this gap, we introduce ReCoQA, a large-scale benchmark of 29,270 real-estate instances featuring machine-verifiable supervision for intermediate steps, including structured intent labels, SQL queries, and API calls. Complementarily, we propose HIRE-Agent, a hierarchical framework instantiating an understand-plan-execute architecture as a strong baseline. By orchestrating a Front-end parser, a planning Supervisor, and execution Specialists, HIRE-Agent effectively integrates heterogeneous evidence. Extensive experiments demonstrate that HIRE-Agent constitutes a strong baseline and substantiates the necessity of hierarchical collaboration for complex, real-world reasoning tasks.

Yindong Zhang, Wenmian Yang, Yiquan Zhang +1

Hong Kong Baptist University · Beijing Normal-Hong Kong Baptist University · Beijing Normal University

LLM Agent OrchestrationAI Agent BenchmarksTool-Using AgentsLLM Tool UseKnowledge-Intensive QA
AskPDF
Apr 20, 2026

GraSP: Graph-Structured Skill Compositions for LLM Agents

Skill ecosystems for LLM agents have matured rapidly, yet recent benchmarks show that providing agents with more skills does not monotonically improve performance -- focused sets of 2-3 skills outperform comprehensive documentation, and excessive skills actually hurt. The bottleneck has shifted from skill availability to skill orchestration: agents need not more skills, but a structural mechanism to select, compose, and execute them with explicit causal dependencies. We propose GraSP, the first executable skill graph architecture that introduces a compilation layer between skill retrieval and execution. GraSP transforms flat skill sets into typed directed acyclic graphs (DAGs) with precondition-effect edges, executes them with node-level verification, and performs locality-bounded repair through five typed operators -- reducing replanning from O(N) to O(d^h). Across ALFWorld, ScienceWorld, WebShop, and InterCode with eight LLM backbones, GraSP outperforms ReAct, Reflexion, ExpeL, and flat skill baselines in every configuration, improving reward by up to +19 points over the strongest baseline while cutting environment steps by up to 41%. GraSP's advantage grows with task complexity and is robust to both skill over-retrieval and quality degradation, confirming that structured orchestration -- not larger skill libraries -- is the key to reliable agent execution.

Tianle Xia, Lingxiang Hu, Yiding Sun +5

Tencent

LLM Agent OrchestrationLLM Agent Workflow OptimizationLLM AgentsLLM Agent Planning
AskPDF
Apr 20, 2026

Co-evolving Agent Architectures and Interpretable Reasoning for Automated Optimization

Automating operations research (OR) with large language models (LLMs) remains limited by hand-crafted reasoning--execution workflows. Complex OR tasks require adaptive coordination among problem interpretation, mathematical formulation, solver selection, code generation, and iterative debugging. To address this limitation, we propose EvoOR-Agent, a co-evolutionary framework for automated optimization. The framework represents agent workflows as activity-on-edge (AOE)-style networks, making workflow topology, execution dependencies, and alternative reasoning paths explicit. On this representation, the framework maintains an architecture graph and evolves a population of reasoning individuals through graph-mediated path-conditioned recombination, multi-granularity semantic mutation, and elitist population update. A knowledge-base-assisted experience-acquisition module further injects reusable OR practices into initialization and semantic variation. Empirical results on heterogeneous OR benchmarks show that the proposed framework consistently improves over zero-shot LLMs, fixed-pipeline OR agents, and representative evolutionary agent frameworks. Case studies and ablation analyses further indicate that explicit architecture evolution and graph-supported reasoning-trajectory search contribute to both performance improvement and structural interpretability. These results suggest that treating agent architectures and reasoning trajectories as evolvable objects provides an effective route toward adaptive and interpretable automated optimization.

Jiahao Huang, Peilan Xu, Xiaoya Nan +1

School of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing 210044, China · Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies, Institute of Cyberspace Security, School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China

Evolutionary OptimizationLarge Language Model-Guided OptimizationLLM Agent OrchestrationLLM Agents
AskPDF
Apr 18, 2026

Graph-of-Agents: A Graph-based Framework for Multi-Agent LLM Collaboration

With an ever-growing zoo of LLMs and benchmarks, the need to orchestrate multiple models for improved task performance has never been more pressing. While frameworks like Mixture-of-Agents (MoA) attempt to coordinate LLMs, they often fall short in terms of (1) selecting relevant agents, (2) facilitating effective intra-agent communication, and (3) integrating responses efficiently. In this work, we propose Graph-of-Agents (GoA), a new graph-based framework for modeling multi-agent LLM communication. Our approach begins with node sampling, selecting only the most relevant agents by leveraging model cards that summarize each model's domain, task specialization, and other characteristics. Next, we construct edges between the selected agents by evaluating their responses against one another to determine relevance ordering. Directed message passing is then performed from highly relevant agents to less relevant ones to enhance their responses, followed by reverse message passing to refine the original responses of the more relevant agents. Finally, the updated responses are aggregated via graph-based pooling (e.g., max or mean pooling) to produce a single, unified answer. We evaluate GoA on diverse multi-domain benchmarks (MMLU, MMLU-Pro, GPQA) and domain-specific benchmarks (MATH, HumanEval, MedMCQA), with an agent pool of 6 LLMs spanning multiple domains. Surprisingly, GoA achieves superior performance using only 3 selected agents, outperforming recent multi-agent LLM baselines that utilize all 6 agents simultaneously. By adopting a graph structure, GoA offers both scalability and effectiveness through structured message passing-positioning it as a strong candidate for navigating the challenges of the ever-growing LLM zoo. Code is available at: https://github.com/UNITES-Lab/GoA.

Sukwon Yun, Jie Peng, Pingzhi Li +5

UNC Chapel Hill · Eigent AI · MIT-IBM Watson AI Lab +1

LLM Agent OrchestrationMulti-Agent CollaborationMulti-Agent CommunicationMulti-LLM Collaboration
AskPDF
Apr 18, 2026

Beyond Task Success: An Evidence-Synthesis Framework for Evaluating, Governing, and Orchestrating Agentic AI

Agentic AI systems plan, use tools, maintain state, and act across multi-step workflows with external effects, meaning trustworthy deployment can no longer be judged by task completion alone. The current literature remains fragmented across benchmark-centered evaluation, standards-based governance, orchestration architectures, and runtime assurance mechanisms. This paper contributes a bounded evidence synthesis across a manually coded corpus of twenty-four recent sources. The core finding is a governance-to-action closure gap: evaluation tells us whether outcomes were good, governance defines what should be allowed, but neither identifies where obligations bind to concrete actions or how compliance can later be proven. To close that gap, the paper introduces three linked artifacts: (1) a four-layer framework spanning evaluation, governance, orchestration, and assurance; (2) an ODTA runtime-placement test based on observability, decidability, timeliness, and attestability; and (3) a minimum action-evidence bundle for state-changing actions. Across sources, evaluation papers identify safety, robustness, and trajectory-level measurement as open gaps; governance frameworks define obligations but omit execution-time control logic; orchestration research positions the control plane as the locus of policy mediation, identity, and telemetry; runtime-governance work shows path-dependent behavior cannot be governed through prompts or static permissions alone; and action-safety studies show text alignment does not reliably transfer to tool actions. A worked enterprise procurement-agent scenario illustrates how these artifacts consolidate existing evidence without introducing new experimental data.

Christopher Koch, Joshua Andreas Wellbrock

Independent Researcher

LLM Agent OrchestrationAI Agent EvaluationAI Agent SafetyAI Agent Governance
AskPDF
Apr 18, 2026

HiveMind: OS-Inspired Scheduling for Concurrent LLM Agent Workloads

When multiple LLM coding agents share a rate-limited API endpoint, they exhibit resource contention patterns analogous to unscheduled OS processes competing for CPU, memory, and I/O. In a motivating incident, 3 of 11 parallel agents died from connection resets and HTTP 502 errors - a 27% failure rate - despite the API having sufficient aggregate capacity to serve all 11 sequentially. We present HIVEMIND, a transparent HTTP proxy that applies five OS-inspired scheduling primitives - admission control, rate-limit tracking, AIMD backpressure with circuit breaking, token budget management, and priority queuing - to eliminate the failure modes caused by uncoordinated parallel execution. The proxy requires zero modifications to existing agent code and supports Anthropic, OpenAI, and local model APIs via auto-detected provider profiles. Our evaluation across seven scenarios (5-50 concurrent agents) shows that uncoordinated agents fail at 72-100% rates under contention, while HIVEMIND reduces failures to 0-18% and eliminates 48-100% of wasted compute. An ablation study reveals that transparent retry - not admission control - is the single most critical primitive, but the primitives are most effective in combination. Real-world validation against Ollama confirms that HIVEMIND adds under 3ms of proxy overhead per request. The system is open-source under the MIT license.

Justice Owusu Agyemang, Jerry John Kponyo, Obed Kwasi Somuah +3

Sperix Labs · VIA Cybersecurity Lab, KNUST · Quantum and Assistive Technologies Lab, KNUST

LLM Inference EfficiencyAI Coding AgentsLLM Agent OrchestrationLLM Agent ReliabilityLLM Request Scheduling
AskPDF
Apr 18, 2026

Harness as an Asset: Enforcing Determinism via the Convergent AI Agent Framework (CAAF)

Large Language Models produce a controllability gap in safety-critical engineering: even low rates of undetected constraint violations render a system undeployable. Current orchestration paradigms suffer from sycophantic compliance, context attention decay, and stochastic oscillation during self-correction. We introduce the Convergent AI Agent Framework (CAAF), which transitions agentic workflows from open-loop generation to closed-loop fail-safe determinism via three pillars: (1) Recursive Atomic Decomposition with physical context firewalls; (2) Harness as an Asset, formalizing domain invariants into machine-readable registries enforced by a deterministic Unified Assertion Interface; and (3) Structured Semantic Gradients with State Locking for monotonic non-regression. This paper makes two core claims. First, an industrialization thesis: once domain invariants are formalized as an executable Harness, the Harness itself becomes a first-class enterprise asset that compounds in value as foundation models commoditize, and CAAF's ability to deliver its reliability on commodity-tier models makes fully self-hosted, on-premises deployment architecturally feasible for regulated sectors where cloud APIs are not an option. Second, an architectural claim supported by ablation: CAAF's three pillars address complementary failure surfaces and none alone closes the controllability gap at commodity cost. The paper contributes entirely at the orchestration and industrialization layer. Evidence across two complementary benchmarks, three-tier UAI ablations, multi-agent baselines, and a closed-source commodity family replicated by two independent open-weight families, is reported in the body.

Tianbao Zhang

Independent Researcher

Safety-Critical ControlLLM GuardrailsLLM Agent OrchestrationLLM Agent Reliability
AskPDF
Apr 18, 2026

Small Model as Master Orchestrator: Learning Unified Agent-Tool Orchestration with Parallel Subtask Decomposition

Multi-agent systems (MAS) demonstrate clear advantages in tackling complex problems by coordinating diverse agents and external tools. However, most existing orchestration methods rely on static workflows or serial agent scheduling, and are further constrained by heterogeneous interface protocols between tools and agents. This leads to high system complexity and poor extensibility. To mitigate these issues, we propose Agent-as-Tool, a unified parallel orchestration paradigm that abstracts both agents and tools into a standardized, learnable action space with protocol normalization and explicit state feedback. Building on this paradigm, we train a lightweight orchestrator, ParaManager, which decouples planning decisions from subtask solving, enabling state-aware parallel subtask decomposition, delegation, and asynchronous execution. For training, we adopt a two-stage ParaManager training pipeline. It improves robustness by incorporating supervised fine-tuning (SFT) trajectories equipped with recovery mechanisms, and further applies reinforcement learning (RL) to achieve an optimal balance among task success, protocol compliance, diversity, and reasoning efficiency. Experiments show that ParaManager achieves strong performance across multiple benchmarks and exhibits robust generalization under unseen model pools.

Wenzhen Yuan, Wutao Xiong, Fanchen Yu +7

Shanghai Jiao Tong University · Sichuan University · Shanghai AI Lab +2

Multi-Agent System OptimizationMulti-Agent OrchestrationLLM Agent OrchestrationLLM Agent TrainingMulti-Agent Planning
AskPDF
Apr 17, 2026

Agentic Frameworks for Reasoning Tasks: An Empirical Study

Recent advances in agentic frameworks have enabled AI agents to perform complex reasoning and decision-making. However, evidence comparing their reasoning performance, efficiency, and practical suitability remains limited. To address this gap, we empirically evaluate 22 widely used agentic frameworks across three reasoning benchmarks: BBH, GSM8K, and ARC. The frameworks were selected from 1,200 GitHub repositories collected between January 2023 and July 2025 and organized into a taxonomy based on architectural design. We evaluated them under a unified setting, measuring reasoning accuracy, execution time, computational cost, and cross-benchmark consistency. Our results show that 19 of the 22 frameworks completed all three benchmarks. Among these, 12 showed stable performance, with mean accuracy of 74.6-75.9%, execution time of 4-6 seconds per task, and cost of 0.14-0.18 cents per task. Poorer results were mainly caused by orchestration problems rather than reasoning limits. For example, Camel failed to complete BBH after 11 days because of uncontrolled context growth, while Upsonic consumed USD 1,434 in one day because repeated extraction failures triggered costly retries. AutoGen and Mastra also exhausted API quotas through iterative interactions that increased prompt length without improving results. We also found a sharp drop in mathematical reasoning. Mean accuracy on GSM8K was 44.35%, compared with 89.80% on BBH and 89.56% on ARC. Overall, this study provides the first large-scale empirical comparison of agentic frameworks for reasoning-intensive software engineering tasks and shows that framework selection should prioritize orchestration quality, especially memory control, failure handling, and cost management.

Zeeshan Rasheed, Abdul Malik Sami, Muhammad Waseem +3

*Faculty of Information Technology and Communication Sciences, Tampere University, Tampere, Finland.

LLM Agent OrchestrationAI Agent EvaluationAI Agent BenchmarksAlgorithmic Reasoning
AskPDF
Apr 15, 2026

Evo-MedAgent: Beyond One-Shot Diagnosis with Agents That Remember, Reflect, and Improve

Tool-augmented large language model (LLM) agents can orchestrate specialist classifiers, segmentation models, and visual question-answering modules to interpret chest X-rays. However, these agents still solve each case in isolation: they fail to accumulate experience across cases, correct recurrent reasoning mistakes, or adapt their tool-use behavior without expensive reinforcement learning. While a radiologist naturally improves with every case, current agents remain static. In this work, we propose Evo-MedAgent, a self-evolving memory module that equips a medical agent with the capacity for inter-case learning at test time. Our memory comprises three complementary stores: (1)~Retrospective Clinical Episodes that retrieve problem-solving experiences from similar past cases, (2)~an Adaptive Procedural Heuristics bank curating priority-tagged diagnostic rules that evolves via reflection, much like a physician refining their internal criteria, and (3)~a Tool Reliability Controller that tracks per-tool trustworthiness. On ChestAgentBench, Evo-MedAgent raises multiple-choice question (MCQ) accuracy from 0.68 to 0.79 on GPT-5-mini, and from 0.76 to 0.87 on Gemini-3 Flash. With a strong base model, evolving memory improves performance more effectively than orchestrating external tools on qualitative diagnostic tasks. Because Evo-MedAgent requires no training, its per-case overhead is bounded by one additional retrieval pass and a single reflection call, making it deployable on top of any frozen model.

Weixiang Shen, Bailiang Jian, Jun Li +6

Technical University of Munich (TUM) · TUM University Hospital · LMU Munich +3

Medical AI AgentsLLM Agent Self-ImprovementLLM Agent MemoryContinual Learning for LLM AgentsLLM Agent OrchestrationMedical Image Analysis
AskPDF
Apr 15, 2026

AIBuildAI: An AI Agent for Automatically Building AI Models

AI models underpin modern intelligent systems, driving advances across science, medicine, finance, and technology. Yet developing high-performing AI models remains a labor-intensive process that requires expert practitioners to iteratively design architectures, engineer representations, implement training pipelines and refine approaches through empirical evaluation. Existing AutoML methods partially alleviate this burden but remain limited to narrow aspects such as hyperparameter optimization and model selection within predefined search spaces, leaving the full development lifecycle largely dependent on human expertise. To address this gap, we introduce AIBuildAI, an AI agent that automatically builds AI models from a task description and training data. AIBuildAI adopts a hierarchical agent architecture in which a manager agent coordinates three specialized sub-agents: a designer for modeling strategy, a coder for implementation and debugging, and a tuner for training and performance optimization. Each sub-agent is itself a large language model (LLM) based agent capable of multi-step reasoning and tool use, enabling end-to-end automation of the AI model development process that goes beyond the scope of existing AutoML approaches. We evaluate AIBuildAI on MLE-Bench, a benchmark of realistic Kaggle-style AI development tasks spanning visual, textual, time-series and tabular modalities. AIBuildAI ranks first on MLE-Bench with a medal rate of 63.1%, outperforming all existing baseline methods and matching the capability of highly experienced AI engineers. These results demonstrate that hierarchical agent systems can automate the full AI model development process from task specification to deployable model, suggesting a pathway toward broadly accessible AI development with minimal human intervention.

Ruiyi Zhang, Peijia Qin, Qi Cao +2

Department of Electrical and Computer Engineering, University of California San Diego, La Jolla, 92093, CA, USA. · Department of Medicine, University of California San Diego, La Jolla, 92093, CA, USA.

Automated Machine LearningMulti-Agent LLMsLLM Agent Orchestration
AskPDF
Apr 15, 2026

Credo: Declarative Control of LLM Pipelines via Beliefs and Policies

Agentic AI systems are becoming commonplace in domains that require long-lived, stateful decision-making in continuously evolving conditions. As such, correctness depends not only on the output of individual model calls, but also on how to best adapt when incorporating new evidence or revising prior conclusions. However, existing frameworks rely on imperative control loops, ephemeral memory, and prompt-embedded logic, making agent behavior opaque, brittle, and difficult to verify. This paper introduces Credo, which represents semantic state as beliefs and regulates behavior using declarative policies defined over these beliefs. This design supports adaptive, auditable, and composable execution through a database-backed semantic control plane. We showcase these concepts in a decision-control scenario, where beliefs and policies declaratively guide critical execution choices (e.g., model selection, retrieval, corrective re-execution), enabling dynamic behavior without requiring any changes to the underlying pipeline code.

Duo Lu, Andrew Crotty, Uğur Çetintemel

Brown University · Northwestern University

LLM Agent OrchestrationLLM AgentsLanguage Model-Based Control
AskPDF
Apr 15, 2026

GeoAgentBench: A Dynamic Execution Benchmark for Tool-Augmented Agents in Spatial Analysis

The integration of Large Language Models (LLMs) into Geographic Information Systems (GIS) marks a paradigm shift toward autonomous spatial analysis. However, evaluating these LLM-based agents remains challenging due to the complex, multi-step nature of geospatial workflows. Existing benchmarks primarily rely on static text or code matching, neglecting dynamic runtime feedback and the multimodal nature of spatial outputs. To address this gap, we introduce GeoAgentBench (GABench), a dynamic and interactive evaluation benchmark tailored for tool-augmented GIS agents. GABench provides a realistic execution sandbox integrating 117 atomic GIS tools, encompassing 53 typical spatial analysis tasks across 6 core GIS domains. Recognizing that precise parameter configuration is the primary determinant of execution success in dynamic GIS environments, we designed the Parameter Execution Accuracy (PEA) metric, which utilizes a "Last-Attempt Alignment" strategy to quantify the fidelity of implicit parameter inference. Complementing this, a Vision-Language Model (VLM) based verification is proposed to assess data-spatial accuracy and cartographic style adherence. Furthermore, to address the frequent task failures caused by parameter misalignments and runtime anomalies, we developed a novel agent architecture, Plan-and-React, that mimics expert cognitive workflows by decoupling global orchestration from step-wise reactive execution. Extensive experiments with seven representative LLMs demonstrate that the Plan-and-React paradigm significantly outperforms traditional frameworks, achieving the optimal balance between logical rigor and execution robustness, particularly in multi-step reasoning and error recovery. Our findings highlight current capability boundaries and establish a robust standard for assessing and advancing the next generation of autonomous GeoAI.

Bo Yu, Cheng Yang, Dongyang Hou +6

School of Geosciences and Info-Physics, Central South University, Changsha, China. · School of Earth Sciences and Spatial Information Engineering, Hunan University of Science and Technology, Xiangtan, China.

LLM Agent EvaluationAI Agent BenchmarksLLM Agent OrchestrationSpatial Reasoning
AskPDF
Apr 14, 2026

AgentSPEX: An Agent SPecification and EXecution Language

Language-model agent systems commonly rely on reactive prompting, in which a single instruction guides the model through an open-ended sequence of reasoning and tool-use steps, leaving control flow and intermediate state implicit and making agent behavior potentially difficult to control. Orchestration frameworks such as LangGraph, DSPy, and CrewAI impose greater structure through explicit workflow definitions, but tightly couple workflow logic with Python, making agents difficult to maintain and modify. In this paper, we introduce AgentSPEX, an Agent SPecification and EXecution Language for specifying LLM-agent workflows with explicit control flow and modular structure, along with a customizable agent harness. AgentSPEX supports typed steps, branching and loops, parallel execution, reusable submodules, and explicit state management, and these workflows execute within an agent harness that provides tool access, a sandboxed virtual environment, and support for checkpointing, verification, and logging. Furthermore, we provide a visual editor with synchronized graph and workflow views for authoring and inspection. We include ready-to-use agents for deep research and scientific research, and we evaluate AgentSPEX on 7 benchmarks. Finally, we show through a user study that AgentSPEX provides a more interpretable and accessible workflow-authoring paradigm than a popular existing agent framework.

Pengcheng Wang, Jerry Huang, Jiarui Yao +7

University of Illinois Urbana-Champaign · Baylor College of Medicine · University of Wisconsin–Madison

LLM Agent OrchestrationLLM Agent Harnesses
AskPDF
Apr 14, 2026

Multi-Agent Object Detection Framework Based on Raspberry Pi YOLO Detector and Slack-Ollama Natural Language Interface

The paper presents design and prototype implementation of an edge based object detection system within the new paradigm of AI agents orchestration. It goes beyond traditional design approaches by leveraging on LLM based natural language interface for system control and communication and practically demonstrates integration of all system components into a single resource constrained hardware platform. The method is based on the proposed multi-agent object detection framework which tightly integrates different AI agents within the same task of providing object detection and tracking capabilities. The proposed design principles highlight the fast prototyping approach that is characteristic for transformational potential of generative AI systems, which are applied during both development and implementation stages. Instead of specialized communication and control interface, the system is made by using Slack channel chatbot agent and accompanying Ollama LLM reporting agent, which are both run locally on the same Raspberry Pi platform, alongside the dedicated YOLO based computer vision agent performing real time object detection and tracking. Agent orchestration is implemented through a specially designed event based message exchange subsystem, which represents an alternative to completely autonomous agent orchestration and control characteristic for contemporary LLM based frameworks like the recently proposed OpenClaw. Conducted experimental investigation provides valuable insights into limitations of the low cost testbed platforms in the design of completely centralized multi-agent AI systems. The paper also discusses comparative differences between presented approach and the solution that would require additional cloud based external resources.

Vladimir Kalušev, Branko Brkljač, Milan Brkljač

The Institute for Artificial Intelligence Research and Development of Serbia Fruškogorska 1, 21000 Novi Sad, Republic of Serbia · Department of Power, Electronic and Telecommunication Engineering Faculty of Technical Sciences, University of Novi Sad Trg Dositeja Obradovića 6, 21000 Novi Sad, Republic of Serbia · Faculty of Finance, Banking and Auditing, Alfa BK University Bulevar maršala Tolbuhina 8, 11070 Belgrade, Republic of Serbia

Multi-Agent OrchestrationLLM Agent OrchestrationEdge AIObject Detection
AskPDF