# Orchestrator: Active Inference for Multi-Agent Systems in Long-Horizon Tasks

arXiv: https://arxiv.org/abs/2509.05651 · HTML: https://mlsanity.com/paper/2509.05651/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 40 of its 56 references that are in the map.
- First posted on arXiv 6 Sep 2025.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 30 of 40)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors](https://mlsanity.com/paper/2308.10848/) (2023)
- [The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity](https://mlsanity.com/paper/2506.06941/) (2025)
- [Actor-Attention-Critic for Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1810.02912/) (2018)
- [AI Agents That Matter](https://mlsanity.com/paper/2407.01502/) (2024)
- [Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks](https://mlsanity.com/paper/1906.08291/) (2019)
- [Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks](https://mlsanity.com/paper/2503.09572/) (2025)
- [TradingAgents: Multi-Agents LLM Financial Trading Framework](https://mlsanity.com/paper/2412.20138/) (2024)
- [More Agents Is All You Need](https://mlsanity.com/paper/2402.05120/) (2024)
- [LLM-Based Multi-Agent Systems for Software Engineering: Literature Review, Vision and the Road Ahead](https://mlsanity.com/paper/2404.04834/) (2024)
- [Deep Decentralized Multi-task Multi-Agent Reinforcement Learning under Partial Observability](https://mlsanity.com/paper/1703.06182/) (2017)
- [Multi-Agent Collaboration via Evolving Orchestration](https://mlsanity.com/paper/2505.19591/) (2025)
- [SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning](https://mlsanity.com/paper/2503.11951/) (2025)
- [Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System](https://mlsanity.com/paper/2410.08115/) (2024)
- [Teaching Language Models to Critique via Reinforcement Learning](https://mlsanity.com/paper/2502.03492/) (2025)
- [AlphaMaze: Enhancing Large Language Models' Spatial Intelligence via GRPO](https://mlsanity.com/paper/2502.14669/) (2025)
- [LLaMAR: Long-Horizon Planning for Multi-Agent Robots in Partially Observable Environments](https://mlsanity.com/paper/2407.10031/) (2024)
- [Adaptive Graph Pruning for Multi-Agent Communication](https://mlsanity.com/paper/2506.02951/) (2025)
- [MAS-ZERO: Designing Multi-Agent Systems with Zero Supervision](https://mlsanity.com/paper/2505.14996/) (2025)
- [AdaSociety: An Adaptive Environment with Social Structures for Multi-Agent Decision-Making](https://mlsanity.com/paper/2411.03865/) (2024)
- [Memory Gym: Towards Endless Tasks to Benchmark Memory Capabilities of Agents](https://mlsanity.com/paper/2309.17207/) (2023)
- [Multi-Agent Coordination via Multi-Level Communication](https://mlsanity.com/paper/2209.12713/) (2022)
- [Optimizing Sequential Multi-Step Tasks with Parallel LLM Agents](https://mlsanity.com/paper/2507.08944/) (2025)
- [HDFlow: Enhancing LLM Complex Problem-Solving with Hybrid Thinking and Dynamic Workflows](https://mlsanity.com/paper/2409.17433/) (2024)
- [On Implementing Autonomous Supply Chains: a Multi-Agent System Approach](https://mlsanity.com/paper/2310.09435/) (2023)
- [Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization](https://mlsanity.com/paper/2406.02016/) (2024)

## Built on by (most cited first, 2 of 2)

- [AnimAgents: Coordinating Multi-Stage Animation Pre-Production with Human-Multi-Agent Collaboration](https://mlsanity.com/paper/2511.17906/) (2025)
- [Alignment in Time: Peak-Aware Orchestration for Long-Horizon Agentic Systems](https://mlsanity.com/paper/2602.17910/) (2026)

## Abstract (arXiv)

Complex, non-linear tasks challenge LLM-enhanced multi-agent systems (MAS) due to partial observability and suboptimal coordination. We propose Orchestrator, a novel MAS framework that leverages attention-inspired self-emergent coordination and reflective benchmarking to optimize global task performance. Orchestrator introduces a monitoring mechanism to track agent-environment dynamics, using active inference benchmarks to optimize system behavior. By tracking agent-to-agent and agent-to-environment interaction, Orchestrator mitigates the effects of partial observability and enables agents to approximate global task solutions more efficiently. We evaluate the framework on a series of maze puzzles of increasing complexity, demonstrating its effectiveness in enhancing coordination and performance in dynamic, non-linear environments with long-horizon objectives.