LLM-Enhanced Deep Reinforcement Learning for Task Offloading in Collaborative Edge Computing

TL;DR

LeDRL is a hybrid framework combining lightweight LLMs and self-attention DRL to improve real-time task offloading in collaborative edge computing, addressing sample inefficiency and decision latency.

Contribution

This work introduces LeDRL, a novel hybrid decision framework that integrates structured LLM prompts with self-attention DRL for enhanced task offloading performance.

Findings

LeDRL achieves over 17% higher success rate than baselines.

LeDRL demonstrates faster convergence and better responsiveness.

LeDRL is robust and feasible on resource-constrained Jetson devices.

Abstract

Collaborative edge computing uses edge nodes in different locations to execute tasks, necessitating dynamic task offloading decisions to maintain low latency and high reliability, especially under unpredictable node failures. Although deep reinforcement learning (DRL) and large language models (LLMs) have shown promise for task offloading, DRL often suffers from high sample inefficiency and local optima, whereas LLMs struggle with real-time decision-making. To address these limitations, we propose \textbf{LeDRL}, a hybrid decision framework that couples a \emph{lightweight LLM} with self-attention-enhanced DRL for real-time task offloading. LeDRL constructs structured, context-aware prompts capturing node status, task semantics, and link dynamics to derive high-level strategy priors. These are selectively processed by a self-attention-based alignment module for context-aware policy optimization. A reflective evaluator distills semantic feedback from past trajectories to guide future prompts, enabling more informative and temporally generalizable LLM queries. Extensive experiments show that LeDRL outperforms baselines in task success rate, convergence speed, and real-time responsiveness across diverse network scales, achieving over 17\% improvement in success rate. Furthermore, we deploy LeDRL on Jetson-based edge devices using our prototype system \textit{CoEdgeSys}, demonstrating its robustness and feasibility under resource constraints. Our code is available at:https://github.com/GalleyG5/LeDRL.git.