Exploring Large Language Model as an Interactive Sports Coach: Lessons from a Single-Subject Half Marathon Preparation

2025-09-30

AI Summary

This study investigates the feasibility and limitations of large language models (LLMs) as longitudinal virtual sports coaches for half-marathon training. Addressing key shortcomings of existing LLM-based coaching—including absence of real-time multimodal sensing, passive responsiveness, insufficient personalization, and weak privacy safeguards—we propose a closed-loop virtual coaching framework integrating natural language interaction, consumer-grade运动 telemetry (e.g., heart rate, cadence), and persistent athlete modeling. The framework enables proactive motivation, dynamic training plan adaptation, and privacy-preserving personalized feedback. Empirical evaluation over two months demonstrated that participants progressed from sustaining 2 km runs to completing a half-marathon (21.1 km at 6′30″/km pace); significant improvements in cadence–heart-rate coupling and运动 efficiency index further confirmed the approach’s safety, efficacy, and sustainability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Computer Vision: Motion & Tracking

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Large language models (LLMs) are emerging as everyday assistants, but their role as longitudinal virtual coaches is underexplored. This two-month single subject case study documents LLM guided half marathon preparation (July-September 2025). Using text based interactions and consumer app logs, the LLM acted as planner, explainer, and occasional motivator. Performance improved from sustaining 2 km at 7min 54sec per km to completing 21.1 km at 6min 30sec per km, with gains in cadence, pace HR coupling, and efficiency index trends. While causal attribution is limited without a control, outcomes demonstrate safe, measurable progress. At the same time, gaps were evident, no realtime sensor integration, text only feedback, motivation support that was user initiated, and limited personalization or safety guardrails. We propose design requirements for next generation systems, persistent athlete models with explicit guardrails, multimodal on device sensing, audio, haptic, visual feedback, proactive motivation scaffolds, and privacy-preserving personalization. This study offers grounded evidence and a design agenda for evolving LLMs from retrospective advisors to closed-loop coaching companions.

Problem

• Investigating LLMs as longitudinal virtual sports coaching assistants
• Addressing limitations in real-time sensor integration and personalization
• Proposing design requirements for closed-loop LLM coaching systems

Innovation

• LLM as longitudinal virtual sports coach
• Text-based interaction with consumer app logs
• Proposed multimodal sensing and proactive feedback