CogGuard: Cognitive and Operational Profiling for Proactive Warning in Edge Intelligent Services

2026-06-13

AI Summary

This work addresses the lack of a generalizable proactive alert mechanism for edge intelligence services across diverse scenarios and the challenges of model alignment on heterogeneous edge clusters, which suffer from high synchronization overhead and load imbalance due to varying input lengths. To tackle these issues, the authors propose CogGuard, a framework that decouples offline large language models to construct cognitive-action profiles and leverages lightweight online language models to predict task success rates with low latency and high efficiency. The approach introduces three key innovations: a static-dynamic reusable profile abstraction, a prefix-aligned KV cache reuse mechanism, and a length-aware distributed fine-tuning strategy. Experiments on educational and operational task datasets demonstrate up to 48% reduction in profile construction time, 19% faster fine-tuning, and a maximum 15.4% decrease in prediction error, achieving mean absolute errors of 13.4 and 5.9 (on a 100-point scale), respectively.

Abstract

Proactive warning is an important capability for edge intelligent services, where the system predicts whether a subject will successfully complete an incoming task under strict latency and privacy constraints. Such prediction depends on both long-term static attributes and short-term dynamic states derived from historical interaction logs. Recent Large Language Models (LLMs) offer strong long-context reasoning for constructing structured profiles from these logs, but existing solutions face two challenges for edge deployment: (1) profiling methods are typically domain-specific and lack a reusable abstraction across service scenarios, and (2) fine-tuning alignment models on heterogeneous edge clusters incurs high synchronization overhead due to the variance in input sequence lengths. To address these challenges, we propose CogGuard, a proactive-warning framework for edge intelligent services. CogGuard decouples offline LLM-based profile construction from online Small Language Model (SLM)-based score prediction through a shared static-dynamic profile-to-score pipeline, and instantiates it in two representative scenarios: educational performance warning and operational task outcome warning. For efficient profile construction, we design scenario-specific profiling methods with prefix-aligned KV-cache reuse to reduce repeated encoding overhead. For edge-side model alignment, we propose a length-aware distributed fine-tuning strategy with contrastive regularization to mitigate workload imbalance on heterogeneous clusters. Experiments on education and operation datasets show that CogGuard reduces profile construction time by up to 48% and distributed fine-tuning time by 19%, while achieving MAEs of 13.4 and 5.9, respectively, on 100-point-scale warning tasks. In the largest educational setting, CogGuard reduces prediction error by 15.4% compared with the strongest baseline.

Problem

• proactive warning
• edge intelligent services
• cognitive profiling
• operational profiling
• heterogeneous edge clusters

Innovation

• edge intelligence
• proactive warning
• language model decoupling
• KV-cache reuse
• length-aware fine-tuning