DIANOIA: Diagnostic Decomposition and Joint Optimization for Multi-Agent Reasoning
Yiming Yang, Zhuoyuan Li, Fanxiang Zeng, Hao Fu, Yue Liu
Why It Matters
What makes this one worth your time
This framework enables practitioners to make informed design choices for multi-agent systems, improving efficiency and effectiveness in various tasks.
DIANOIA provides a structured approach to optimize multi-agent systems by diagnosing performance bottlenecks.
Summary
The paper introduces DIANOIA, a diagnostic framework for multi-agent reasoning that decomposes performance into measurable components, allowing for targeted optimization based on identified bottlenecks.
Key contributions
- Introduction of a three-channel decomposition for multi-agent reasoning.
- Development of a diagnostic protocol that identifies performance bottlenecks.
- Empirical validation of the proposed method on multiple benchmarks, demonstrating significant performance improvements.
Notable insights
- The three-channel decomposition into coverage, fidelity, and synthesis offers a novel way to analyze multi-agent performance.
- The diagnostic protocol allows for empirical measurement and targeted resource allocation based on identified bottlenecks.
Possible limitations
- Not stated in the abstract.
Abstract
arXiv:2602.08586v3 Announce Type: replace Abstract: Multi-agent LLM systems consistently outperform single-agent baselines, yet practitioners still cannot predict which design works for a new task or diagnose why one fails. We argue this gap persists largely because the field lacks a diagnostic framework with measurable primitives and testable predictions. We introduce \textbf{DIANOIA}, a three-channel decomposition of multi-agent reasoning gain into coverage, fidelity, and synthesis, each of which is empirically measurable. From this decomposition, we derive a diagnostic protocol that identifies the bottleneck channels for any given task. We instantiate the protocol as a multi-agent system whose three components mirror the channels: role-diverse proposers for coverage, execution-grounded verification for fidelity, and iterative synthesis. On GSM8K, AIME-2025, MBPP, and BFCL-SP, our method outperforms strong multi-agent baselines under matched token budgets, dominating the Pareto frontier on MBPP at $\sim$$5{\times}$ token savings and reaching $+4.6$pp at matched cost. On every benchmark, the protocol picks the right bottleneck channels; the system we built around it leads across models. We release code, adapters, diagnostic metrics, and a Claude Code skill at https://anonymous.4open.science/r/DIANOIA4MAS. DIANOIA reframes multi-agent design as channel-aware resource allocation: diagnose which channel is the bottleneck for your task, then invest tokens accordingly.