When Single-Agent with Skills Replace Multi-Agent Systems and When They Fail

• URL: http://arxiv.org/abs/2601.04748v2
• Date: Wed, 14 Jan 2026 07:18:10 GMT
• Title: When Single-Agent with Skills Replace Multi-Agent Systems and When They Fail
• Authors: Xiaoxiao Li,
• Abstract summary: Multi-agent AI systems have proven effective for complex reasoning.<n>Can we achieve similar modularity benefits with a single agent that selects from a library of skills?<n>We investigate the scaling behavior of skill selection and observe a striking pattern.<n>We find evidence that semantic confusability among similar skills, rather than library size alone, plays a central role in this degradation.
• Score: 40.69885101060645
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent AI systems have proven effective for complex reasoning. These systems are compounded by specialized agents, which collaborate through explicit communication, but incur substantial computational overhead. A natural question arises: can we achieve similar modularity benefits with a single agent that selects from a library of skills? We explore this question by viewing skills as internalized agent behaviors. From this perspective, a multi-agent system can be compiled into an equivalent single-agent system, trading inter-agent communication for skill selection. Our preliminary experiments suggest this approach can substantially reduce token usage and latency while maintaining competitive accuracy on reasoning benchmarks. However, this efficiency raises a deeper question that has received little attention: how does skill selection scale as libraries grow? Drawing on principles from cognitive science, we propose that LLM skill selection exhibits bounded capacity analogous to human decision-making. We investigate the scaling behavior of skill selection and observe a striking pattern. Rather than degrading gradually, selection accuracy remains stable up to a critical library size, then drops sharply, indicating a phase transition reminiscent of capacity limits in human cognition. Furthermore, we find evidence that semantic confusability among similar skills, rather than library size alone, plays a central role in this degradation. This perspective suggests that hierarchical organization, which has long helped humans manage complex choices, may similarly benefit AI systems. Our initial results with hierarchical routing support this hypothesis. This work opens new questions about the fundamental limits of semantic-based skill selection in LLMs and offers a cognitive-grounded framework and practical guidelines for designing scalable skill-based agents.

Related papers

• AI Agents for Inventory Control: Human-LLM-OR Complementarity [12.448705668487852]
  Large language models (LLMs) have generated interest in AI agents that can reason flexibly and incorporate rich contextual signals.<n>We study how OR algorithms, LLMs, and humans can interact and complement each other in a multi-period inventory control setting.<n>We show that, on average, human-AI teams achieve higher profits than either humans or AI agents operating alone.
  arXiv  Detail & Related papers  (2026-02-13T05:23:46Z)
• Towards a Science of Collective AI: LLM-based Multi-Agent Systems Need a Transition from Blind Trial-and-Error to Rigorous Science [70.3658845234978]
  Large Language Models (LLMs) have greatly extended the capabilities of Multi-Agent Systems (MAS)<n>Despite this rapid progress, the field still relies heavily on empirical trial-and-error.<n>This bottleneck stems from the ambiguity of attribution.<n>We propose a factor attribution paradigm to systematically identify collaboration-driving factors.
  arXiv  Detail & Related papers  (2026-02-05T04:19:52Z)
• AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent [57.10083973844841]
  AgentArk is a novel framework to distill multi-agent dynamics into the weights of a single model.<n>We investigate three hierarchical distillation strategies across various models, tasks, scaling, and scenarios.<n>By shifting the burden of computation from inference to training, the distilled models preserve the efficiency of one agent while exhibiting strong reasoning and self-correction performance of multiple agents.
  arXiv  Detail & Related papers  (2026-02-03T19:18:28Z)
• Can an Individual Manipulate the Collective Decisions of Multi-Agents? [53.01767232004823]
  M-Spoiler is a framework that simulates agent interactions within a multi-agent system to generate adversarial samples.<n>M-Spoiler introduces a stubborn agent that actively aids in optimizing adversarial samples.<n>Our findings confirm the risks posed by the knowledge of an individual agent in multi-agent systems.
  arXiv  Detail & Related papers  (2025-09-20T01:54:20Z)
• Is the `Agent' Paradigm a Limiting Framework for Next-Generation Intelligent Systems? [0.0]
  The concept of the 'agent' has profoundly shaped Artificial Intelligence (AI) research.<n>This paper critically re-evaluates the necessity and optimality of this agent-centric paradigm.
  arXiv  Detail & Related papers  (2025-09-13T16:11:27Z)
• A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems [93.8285345915925]
  Reasoning is a fundamental cognitive process that enables logical inference, problem-solving, and decision-making.<n>With the rapid advancement of large language models (LLMs), reasoning has emerged as a key capability that distinguishes advanced AI systems.<n>We categorize existing methods along two dimensions: (1) Regimes, which define the stage at which reasoning is achieved; and (2) Architectures, which determine the components involved in the reasoning process.
  arXiv  Detail & Related papers  (2025-04-12T01:27:49Z)
• Kolb-Based Experiential Learning for Generalist Agents with Human-Level Kaggle Data Science Performance [81.05882480184587]
  We propose a computational framework of Kolb's learning cycle with Vygotsky's ZPD for autonomous agents.<n>Agent K is the 1st AI system to successfully integrate Kolb- and Vygotsky-inspired human cognitive learning.<n>With 9 gold, 8 silver, and 12 bronze medals level performance - including 4 gold and 4 silver on prize-awarding competitions - Agent K is the 1st AI system to successfully integrate Kolb- and Vygotsky-inspired human cognitive learning.
  arXiv  Detail & Related papers  (2024-11-05T23:55:23Z)
• Inverse Factorized Q-Learning for Cooperative Multi-agent Imitation Learning [13.060023718506917]
  imitation learning (IL) is a problem of learning to mimic expert behaviors from demonstrations in cooperative multi-agent systems. We introduce a novel multi-agent IL algorithm designed to address these challenges. Our approach enables the centralized learning by leveraging mixing networks to aggregate decentralized Q functions.
  arXiv  Detail & Related papers  (2023-10-10T17:11:20Z)
• Priors, Hierarchy, and Information Asymmetry for Skill Transfer in Reinforcement Learning [18.865535706610522]
  We show the crucial expressivity-transferability trade-off of skills across sequential tasks controlled by information asymmetry. We introduce Attentive Priors for Expressive and Transferable Skills (APES) Unlike existing approaches, APES automates the choice of asymmetry by learning it in a data-driven, domain-dependent, way.
  arXiv  Detail & Related papers  (2022-01-20T11:12:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.