DIAMOND: An LLM-Driven Agent for Context-Aware Baseball Highlight Summarization
Jeonghun Kang, Soonmok Kwon, Joonseok Lee, Byung-Hak Kim
One-line Summary
DIAMOND is an LLM-driven agent that enhances baseball highlight summarization by integrating sabermetrics with natural language reasoning, significantly improving performance over traditional methods.
Plain-language Overview
Traditional methods for summarizing baseball highlights often miss the strategic and narrative elements of the game. DIAMOND is a new tool that uses advanced analytics and language models to better capture the important moments in baseball games. By combining statistical measures of play importance with an understanding of the game's storyline, DIAMOND provides summaries that are both accurate and engaging. Tested on Korean Baseball games, it significantly outperforms existing methods, suggesting a promising new approach for sports highlight summarization.