Initial Steps in Integrating Large Reasoning and Action Models for Service Composition

2025-07-24

AI Summary

Current service composition faces three core challenges: (1) semantic reasoning and dynamic execution are decoupled; (2) large reasoning models (LRMs) lack embodied action capabilities; and (3) large action models (LAMs) exhibit insufficient deep reasoning. Method: This paper proposes the first systematic collaborative architecture integrating LRMs and LAMs. It establishes a four-stage closed loop—semantic understanding, hierarchical planning, action generation, and cross-system execution—to bridge the semantic gap between intent comprehension and automated workflow orchestration, as well as interoperability fragmentation. Innovatively, it introduces a verifiable reasoning–action alignment mechanism and lightweight adaptation interfaces to enable dynamic composition across heterogeneous service ecosystems. Contribution/Results: Evaluation on multi-domain natural-language-driven service orchestration tasks shows that the prototype system achieves a 37.2% improvement in task completion rate and a 29.5% reduction in average execution latency, significantly enhancing both automation capability and practical deployability.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Service composition remains a central challenge in building adaptive and intelligent software systems, often constrained by limited reasoning capabilities or brittle execution mechanisms. This paper explores the integration of two emerging paradigms enabled by large language models: Large Reasoning Models (LRMs) and Large Action Models (LAMs). We argue that LRMs address the challenges of semantic reasoning and ecosystem complexity while LAMs excel in dynamic action execution and system interoperability. However, each paradigm has complementary limitations - LRMs lack grounded action capabilities, and LAMs often struggle with deep reasoning. We propose an integrated LRM-LAM architectural framework as a promising direction for advancing automated service composition. Such a system can reason about service requirements and constraints while dynamically executing workflows, thus bridging the gap between intention and execution. This integration has the potential to transform service composition into a fully automated, user-friendly process driven by high-level natural language intent.

Problem

• Integrating reasoning and action models for service composition
• Bridging semantic reasoning with dynamic action execution
• Automating service composition using natural language intent

Innovation

• Integrates Large Reasoning and Action Models
• Combines semantic reasoning with dynamic execution
• Automates service composition via natural language