{"ID":2868608,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.15602","arxiv_id":"2509.15602","title":"TennisTV: Do Multimodal Large Language Models Understand Tennis Rallies?","abstract":"Multimodal large language models (MLLMs) excel at general video understanding but struggle with fast, high-frequency sports like tennis, where rally clips are short yet information-dense. To systematically evaluate MLLMs in this challenging domain, we present TennisTV, the first and most comprehensive benchmark for tennis video understanding. TennisTV models each rally as a temporal-ordered sequence of consecutive stroke events, using automated pipelines for filtering and question generation. It covers 8 tasks from the stroke level to the rally level and includes 2527 human-verified questions. Evaluating 17 representative MLLMs, we provide the first systematic assessment of tennis video understanding. Results yield two key insights: (i) frame-sampling density should be tailored and balanced across tasks, and (ii) improving temporal grounding is essential for stronger reasoning.","short_abstract":"Multimodal large language models (MLLMs) excel at general video understanding but struggle with fast, high-frequency sports like tennis, where rally clips are short yet information-dense. To systematically evaluate MLLMs in this challenging domain, we present TennisTV, the first and most comprehensive benchmark for ten...","url_abs":"https://arxiv.org/abs/2509.15602","url_pdf":"https://arxiv.org/pdf/2509.15602v5","authors":"[\"Zhongyuan Bao\",\"Lejun Zhang\"]","published":"2025-09-19T05:08:05Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}