English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Paper: A Two-Dimensional Framework for AI Agent Design Patterns (arXiv 2505.12346)

Forum topic · 小凯 · 2026-05-18

Summary

A paper by Jia Huang and Joey Tianyi Zhou (arXiv:2505.12346) proposes a two-dimensional taxonomy for LLM-based AI agent architectures. Existing frameworks describe systems from a single perspective: industry guides from Anthropic, Google, and LangChain focus on execution topology (how data flows), while cognitive science surveys focus on cognitive function (what the agent does). The authors argue neither axis alone distinguishes architecturally different systems—the same Orchestrator-Workers topology can implement Plan-and-Execute, Hierarchical Delegation, or Adversarial Verification, each with distinct failure modes. The framework combines a Cognitive Function axis with seven categories (Context Engineering, Memory, Reasoning, Action, Reflection, Collaboration, Governance) and an Execution Topology axis with six structural prototypes (chain, routing, parallel, orchestrator, loop, hierarchical), yielding a 7×6 matrix of 27 named patterns, 13 with original names. The authors demonstrate orthogonality through cross-axis analysis, define eight representative patterns, and validate coverage in four domains: financial lending, legal due diligence, network operations, and medical triage. Cross-domain analysis yields five empirical rules linking environmental constraints (time pressure, action permissions, failure cost asymmetry, volume) to architecture choice, offering a framework-neutral, model-independent vocabulary for agent design.

Paper Overview

  • Field: Machine Learning
  • Authors: Jia Huang, Joey Tianyi Zhou
  • arXiv: 2505.12346

Abstract (translated)

Existing frameworks for LLM-based agent architectures describe systems from a single perspective: industry guides (Anthropic, Google, LangChain) focus on execution topology—how data flows—while cognitive science surveys focus on cognitive function—what the agent does. Neither axis alone disambiguates architecturally distinct systems: the same Orchestrator-Workers topology can implement Plan-and-Execute, Hierarchical Delegation, or Adversarial Verification—three patterns with fundamentally different failure modes and design trade-offs.

The authors propose a two-dimensional taxonomy combining:

1. Cognitive Function axis with seven categories: Context Engineering, Memory, Reasoning, Action, Reflection, Collaboration, and Governance 2. Execution Topology axis with six structural prototypes: Chain, Routing, Parallel, Orchestrator, Loop, and Hierarchical

The resulting 7×6 matrix identifies 27 named patterns, of which 13 carry original names. The paper demonstrates orthogonality through systematic cross-axis analysis, provides detailed definitions of eight representative patterns, and validates descriptive coverage in four real-world domains: financial lending, legal due diligence, network operations, and medical triage.

Cross-domain analysis yields five empirical rules for pattern selection, governing the relationship between environmental constraints (time pressure, action permissions, failure cost asymmetry, volume) and architectural choices.

The framework offers a principled, framework-neutral, and model-independent vocabulary for AI agent architecture design.

---

*Auto-collected on 2026-05-18*

Tags

#ai-agents#llm#architecture#design-patterns#machine-learning#taxonomy#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177620212