STEPS: Semantic-Contract-Guided Scheduling for LLM-Assisted Natural-Language-Driven Edge AI Services

2026-06-08

AI Summary

This work addresses the semantic gap between users’ imprecise natural language service requests and the numeric constraints required for edge resource scheduling, which hinders accurate mapping into resource-constrained decision-making. To bridge this gap, the authors propose STEPS, a novel framework that introduces “semantic contracts” as an executable interface between natural language and edge orchestration. Leveraging large language models, STEPS parses user requests to generate semantic contracts encoding service preferences, fulfillment boundaries, and semantic uncertainty. These contracts then guide a potential game-theoretic model that jointly optimizes node selection, computational resource allocation, and bandwidth scheduling. Theoretical analysis proves the existence of a pure-strategy Nash equilibrium, ensuring convergence and stability. Experiments demonstrate that STEPS significantly improves contract fulfillment rates, reduces service loss, and exhibits strong robustness and adaptability under ambiguous user requests and non-stationary edge environments.

Abstract

Networked AI services are increasingly delivered through edge infrastructures to support latency-sensitive applications. Edge scheduling is critical for deciding where and how AI services are executed under limited communication and computing resources. Existing frameworks usually assume that requirements are given as numerical constraints, such as latency bounds, energy budgets, or cost limits. In practice, users often express expectations through ambiguous natural language, creating a gap between user intent and resource constrained scheduling. To bridge this gap, we propose semantic-contract-guided edge potential scheduling (STEPS), a natural language driven scheduling framework for LLM assisted edge AI services. STEPS introduces semantic contracts as executable interfaces between user-side semantics and edge-side decision making. An LLM assisted semantic parser extracts service levels and confidence scores, which are converted into service preferences, fulfillment bounds, and semantic uncertainty. Based on these contracts, STEPS formulates edge scheduling as a contract-guided potential game that jointly determines execution-node selection, computing-resource provisioning, and bandwidth allocation. It also builds feedback signals from semantic request drift, fulfillment drift, fulfillment pressure, and admission pressure to adjust semantic admission, contract conservativeness, and edge coordination. We characterize the exact potential game structure, establish pure strategy Nash equilibrium existence, and prove convergence and stability properties. Experiments show that STEPS improves semantic contract fulfillment, reduces contract guided service loss, and maintains robust adaptation under ambiguous requests and non-stationary edge environments.

Problem

• edge AI scheduling
• natural language intent
• semantic gap
• resource-constrained scheduling
• user intent interpretation

Innovation

• semantic contract
• LLM-assisted scheduling
• edge AI
• potential game
• natural language interface