# Coordination as an Architectural Layer for LLM-Based Multi-Agent Systems

arXiv: https://arxiv.org/abs/2605.03310 · HTML: https://mlsanity.com/paper/2605.03310/

## 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 14 of its 35 references that are in the map.
- First posted on arXiv 5 May 2026.

## 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, 14 of 14)

- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [ForecastBench: A Dynamic Benchmark of AI Forecasting Capabilities](https://mlsanity.com/paper/2409.19839/) (2024)
- [Approaching Human-Level Forecasting with Language Models](https://mlsanity.com/paper/2402.18563/) (2024)
- [LLM-Coordination: Evaluating and Analyzing Multi-agent Coordination Abilities in Large Language Models](https://mlsanity.com/paper/2310.03903/) (2023)
- [Wisdom of the Silicon Crowd: LLM Ensemble Prediction Capabilities Rival Human Crowd Accuracy](https://mlsanity.com/paper/2402.19379/) (2024)
- [Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate](https://mlsanity.com/paper/2509.05396/) (2025)
- [Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline](https://mlsanity.com/paper/2601.12307/) (2026)
- [Prediction Arena: Benchmarking AI Models on Real-World Prediction Markets](https://mlsanity.com/paper/2604.07355/) (2026)
- [On the Reliability Limits of LLM-Based Multi-Agent Planning](https://mlsanity.com/paper/2603.26993/) (2026)
- [Foresight Arena: An On-Chain Benchmark for Evaluating AI Forecasting Agents](https://mlsanity.com/paper/2605.00420/) (2026)
- [Semantic Consensus: Process-Aware Conflict Detection and Resolution for Enterprise Multi-Agent LLM Systems](https://mlsanity.com/paper/2604.16339/) (2026)
- [A Declarative Language for Building And Orchestrating LLM-Powered Agent Workflows](https://mlsanity.com/paper/2512.19769/) (2025)

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

- [Nous: An Attempt to Extract and Inject the Cognition Behind Prediction-Market Behavior](https://mlsanity.com/paper/2606.13038/) (2026)
- [Claim Plane: Reliability Gains and the Limits of Selective Concurrency for Parallel Coding Agents: A 30-Pair, Three-Seed Confirmatory Study of Deterministic Pre-Write Admission](https://mlsanity.com/paper/2608.00947/) (2026)

## Abstract (arXiv)

Multi-agent LLM systems fail in production at rates between 41% and 87%, mostly due to coordination defects rather than base-model capability. Existing responses split between cataloguing failure modes empirically and shipping declarative orchestration frameworks as engineering tools; neither delivers a principled mapping from coordination configuration to predictable failure-mode signature. We argue that coordination should be treated as a configurable architectural layer, separable from agent logic and from information access, enabling architectural reasoning rather than only engineering productivity. We instantiate this with an information-controlled design on prediction markets: a single LLM, fixed tools, fixed per-call output cap, and fixed prompt template across five reference coordination configurations, with total compute per question treated as an endogenous architectural output. The Murphy decomposition of the Brier score separates calibration from discriminative power, so configurations leave distinguishable signatures even when aggregate scores coincide. On 100 Polymarket binary markets resolved after the model's training cutoff (claude-opus-4-6) we report Murphy signatures, a cost-quality Pareto frontier, category-conditioned analysis, and a bootstrap power-projection. Three of five pre-specified predictions are upheld in direction; two configurations dominate the Pareto frontier within this regime; exploratory bootstrap intervals separate consensus alignment from others, though pairwise tests do not survive Bonferroni correction at n=100. We also deploy the same configurations as live agents on Foresight Arena under web-search-enabled conditions, as an on-chain replication channel accumulating in parallel. Harness, trace dataset, and production agents are released. We position this as a methodology-validating first instantiation, not a general cross-model claim.