LATIDIA · Investigación

GameCommBench: un punto de referencia unificado y una evaluación consciente de tipos para comentarios de juegos generados por IA

arXiv:2610.11129v1 Tipo de anuncio: nuevo Resumen: El comentario del juego es una tarea de generación abierta que requiere percepción multimodal, razonamiento estratégico y conocimiento contextual. Comentario de juego generado por IA existente (AI-G)

La noticia

arXiv:2610.11129v1 Announce Type: new Abstract: Game commentary is an open-ended generation task requiring multimodal perception, strategic reasoning, and contextual knowledge. Existing AI-Generated Game Commentary (AI-GGC) studies remain fragmented across games, modalities, and evaluation protocols, while overlap-based or holistic evaluators fail to capture the functional heterogeneity of commentary. We introduce \textsc{GameCommBench}, a unified benchmark spanning board games, sports, and esports, with commentary aligned to heterogeneous game contexts and annotated by commentary type. We further propose Type-Aware Commentary Evaluation (TACE), a structured framework for evaluating different types of commentary. We then validate TACE for reliability and human agreement, and use it to benchmark representative AI commentators. Results reveal non-uniform capability profiles, with