DIAMOND: An LLM-Driven Agent for Context-Aware Baseball Highlight Summarization

2025-06-03

AI Summary

Traditional baseball highlight detection methods—such as Win Probability Added (WPA) ranking or vision-based event detection—struggle to jointly capture strategic depth and narrative coherence, while manual annotation remains costly and unscalable. This paper proposes an LLM-driven, context-aware summarization agent that introduces a modular architecture coupling interpretable advanced statistics (WPA, Leverage Index, Win Expectancy) with natural language inference to jointly model importance and storytelling. The method integrates multi-dimensional sports context, optimized prompt engineering, and an interpretable decision-making mechanism. Evaluated on five KBO games, it achieves an F1-score of 84.8%, outperforming the WPA baseline by 41.9 percentage points and significantly surpassing existing statistical and commercial systems.

Technology Category

• Natural Language Processing: Summarization
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Data Visualization & Summarization

Application Category

• 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: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Traditional approaches -- such as Win Probability Added (WPA)-based ranking or computer vision-driven event detection -- can identify scoring plays but often miss strategic depth, momentum shifts, and storyline progression. Manual curation remains the gold standard but is resource-intensive and not scalable. We introduce DIAMOND, an LLM-driven agent for context-aware baseball highlight summarization that integrates structured sports analytics with natural language reasoning. DIAMOND leverages sabermetric features -- Win Expectancy, WPA, and Leverage Index -- to quantify play importance, while an LLM module enhances selection based on contextual narrative value. This hybrid approach ensures both quantitative rigor and qualitative richness, surpassing the limitations of purely statistical or vision-based systems. Evaluated on five diverse Korean Baseball Organization League games, DIAMOND improves F1-score from 42.9% (WPA-only) to 84.8%, outperforming both commercial and statistical baselines. Though limited in scale, our results highlight the potential of modular, interpretable agent-based frameworks for event-level summarization in sports and beyond.

Problem

• Identifying strategic depth and momentum shifts in baseball highlights
• Overcoming resource-intensive manual curation for scalable summarization
• Integrating analytics with narrative context for richer highlight selection

Innovation

• LLM-driven agent for context-aware summarization
• Integrates sports analytics with natural language reasoning
• Hybrid approach combining quantitative and qualitative analysis