aiqbench analysis
DIANOIA introduces a diagnostic framework that decomposes multi-agent LLM reasoning into measurable primitives—coverage, fidelity, and synthesis—to optimize token allocation across specific task bottlenecks. By empirically validating this approach on complex benchmarks like AIME-2025 and MBPP, the research demonstrates significant efficiency gains without sacrificing performance. This work addresses a critical gap in the field by providing actionable metrics for practitioners to predict and improve multi-agent system designs.
DIANOIA: Diagnostic Decomposition and Joint Optimization for Multi-Agent Reasoning
arXiv:2602.08586v4 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$$5{\times}$ token savings and reaching $+4.6$pp at matched cost. On every benchmark, the protocol picks the right bottleneck channels; the system we built around it leads across models. We release code, adapters, diagnostic metrics, and a Claude Code skill at https://anonymous.4open.science/r/DIANOIA4MAS. DIANOIA reframes multi-agent design as channel-aware resource allocation: diagnose which channel is the bottleneck for your task, then invest tokens accordingly.
Abstract:Multi-agent LLM systems consistently outperform single-agent baselines, yet practitioners still cannot predict which design works for a new task or diagnose why one fails. We argue this gap persists largely because the field lacks a diagnostic framework with measurable primitives and testable predictions. We introduce \textbf{DIANOIA}, a three-channel decomposition of multi-agent reasoning gain into coverage, fidelity, and synthesis, each of which is empirically measurable. From this decomposition, we derive a diagnostic protocol that identifies the bottleneck channels for any given task. We instantiate the protocol as a multi-agent system whose three components mirror the channels: role-diverse proposers for coverage, execution-grounded verification for fidelity, and iterative synthesis. On GSM8K, AIME-2025, MBPP, and BFCL-SP, our method outperforms strong multi-agent baselines under matched token budgets, dominating the Pareto frontier on MBPP at $\sim$$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 this https URL. DIANOIA reframes multi-agent design as channel-aware resource allocation: diagnose which channel is the bottleneck for your task, then invest tokens accordingly.
Artificial Intelligence (cs.AI)
arXiv:2602.08586 [cs.AI]
(or arXiv:2602.08586v4 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2602.08586 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 285 words
Related coverage
[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
Latent Space · Sep 23, 2026
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026
MedPrune: Topology-Efficient Multimodal Multi-Agent Communication Evolution for Medical VQA Tasks
arXiv · Oct 6, 2026
SRPO: Setwise Relative Policy Optimization for Multi-Agent LLMs
arXiv · Sep 9, 2026