mlsanityData to 14 Oct 2026

Difficulty-Aware Agentic Orchestration for Query-Specific Multi-Agent Workflows

arXiv 2509.11079 ↗First posted 14 Sep 2025cs.AIcited by 27

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Large Language Model (LLM)-based agentic systems have shown strong capabilities across various tasks. However, existing multi-agent frameworks often rely on static or task-level workflows, which either over-process simple queries or underperform on complex ones, while also neglecting the efficiency-performance trade-offs across heterogeneous LLMs. To address these limitations, we propose Difficulty-Aware Agentic Orchestration (DAAO), which can dynamically generate query-specific multi-agent workflows guided by predicted query difficulty. DAAO comprises three interdependent modules: a variational autoencoder (VAE) for difficulty estimation, a modular operator allocator, and a cost- and performance-aware LLM router. A self-adjusting policy updates difficulty estimates based on workflow success, enabling simpler workflows for easy queries and more complex strategies for harder ones. Experiments on six benchmarks demonstrate that DAAO surpasses prior multi-agent systems in both accuracy and inference efficiency, validating its effectiveness for adaptive, difficulty-aware reasoning.

Key facts. Cited by 27 papers in the map (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 33 of its 68 references that are in the map. First posted on arXiv 14 Sep 2025.

Builds on

33 of 68 references.

References · 33

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  7. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  8. 2021Program Synthesis with Large Language Modelscited by 3,637
  9. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  10. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  11. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
  12. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
21 more
  1. 2016Neural Architecture Search with Reinforcement Learningcited by 1,499
  2. 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,463
  3. 2023Mind2Web: Towards a Generalist Agent for the Webcited by 1,242
  4. 2023GAIA: a benchmark for General AI Assistantscited by 1,128
  5. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 1,011
  6. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 832
  7. 2023LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusioncited by 564
  8. 2022Complexity-Based Prompting for Multi-Step Reasoningcited by 388
  9. 2024Data Interpreter: An LLM Agent For Data Sciencecited by 225
  10. 2025Multi-agent Architecture Search via Agentic Supernetcited by 179
  11. 2024GraphRouter: A Graph-based Router for LLM Selectionscited by 165
  12. 2024EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithmscited by 113
  13. 2024DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solvingcited by 95
  14. 2025MasRouter: Learning to Route LLMs for Multi-Agent Systemscited by 88
  15. 2025GRPO-LEAD: A Difficulty-Aware Reinforcement Learning Approach for Concise Mathematical Reasoning in Language Modelscited by 82
  16. 2024Harder Tasks Need More Experts: Dynamic Routing in MoE Modelscited by 72
  17. 2024AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitionscited by 70
  18. 2024Archon: An Architecture Search Framework for Inference-Time Techniquescited by 60
  19. 2025How Difficulty-Aware Staged Reinforcement Learning Enhances LLMs' Reasoning Capabilities: A Preliminary Experimental Studycited by 14
  20. 2025Know When to Explore: Difficulty-Aware Certainty as a Guide for LLM Reinforcement Learningcited by 12
  21. 2024Automating Customer Needs Analysis: A Comparative Study of Large Language Models in the Travel Industrycited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 27.

27 papers.

Sort:

By area:

  1. 2026From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agentscited by 24
  2. 2026Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systemscited by 20
  3. 2026Sustainable Hybrid Document-Routed Retrieval for Financial RAG: Resolving the Robustness-Precision Trade-offcited by 13
  4. 2026HARBOR: Automated Harness Optimizationcited by 11
  5. 2026Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actionscited by 6
  6. 2026Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priorscited by 6
  7. 2026SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestrationcited by 6
  8. 2026ZEBRA: Zero-shot Budgeted Resource Allocation for LLM Orchestrationcited by 2
  9. 2026AgentRouter: Heterogeneous Model Routing for Cost-Optimal Multi-Step Agentic Workflowscited by 2
  10. 2026Heterogeneous Scientific Foundation Model Collaborationcited by 1
  11. 2026Retrieval-Conditioned Topology Selection with Provable Budget Conservation for Multi-Agent Code Generationcited by 1
  12. 2026LEMON: Learning Executable Multi-Agent Orchestration via Counterfactual Reinforcement Learningcited by 1
15 more
  1. 2026Learning to Hand Off: Provably Convergent Workflow Learning under Interface Constraintscited by 1
  2. 2026Shared Prefixes, Better Credit: Adaptive Routing for Multi-Agent Reasoningcited by 1
  3. 2026FlowScout: From Execution Feedback to Reliable Tool-Using Agent Workflowscited by 1
  4. 2026n-Musketeers: Reinforcement Learning Shapes Collaboration Among Language Modelscited by 0
  5. 2026OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Structured Agentscited by 0
  6. 2026Compiling Deterministic Structure into SLM Harnessescited by 0
  7. 2026FlowCompile: An Optimizing Compiler for Structured LLM Workflowscited by 0
  8. 2026Augment Engineering: A Methodology for Multi-Tool AI Orchestration Across Professional Domainscited by 0
  9. 2026When Does LLM Orchestration Pay Off? A Controlled Evaluation of Accuracy, Cost, and Task Difficultycited by 0
  10. 2026EASy: Towards Efficient LLM-Based Agentic Systemcited by 0
  11. 2026Right-Sizing LLM-Agent Decomposition in VAT Determination: A Pilot Controlled Sweepcited by 0
  12. 2026AtomRec: Evolving Atomic Memory for Agentic Recommendationcited by 0
  13. 2026Pay for the Fault, Not the Flow: Label-Free In-Flow Multi-Agent Workflow Optimizationcited by 0
  14. 2026CB-Orchestrator: Adaptive Workflow Optimization for LLM Agents via Contextual Banditscited by 0
  15. 2026TurnCost: Turns, Not Observations: Cache-Aware Loop Engineering for Coding Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026FlowBank: Query-Adaptive Agentic Workflows Optimization through Precompute-and-Reuse
  2. 2026Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline
  3. 2026OOPMAS: Object-Oriented Multi-Agent Systems for Query-Level Workflow Generation
  4. 2026AgentFactory: Towards Automated Agentic System Design and Optimization
  5. 2025Unifying Language Agent Algorithms with Graph-based Orchestration Engine for Reproducible Agent Research
  6. 2025Parallelism Meets Adaptiveness: Scalable Documents Understanding in Multi-Agent LLM Systems
  7. 2025PAACE: A Plan-Aware Automated Agent Context Engineering Framework
  8. 2025Adaptive Multi-Agent Reasoning via Automated Workflow Generation