一作SIGIR 2026Oral· Information Retrieval (Full)
Generation-Augmented Video Corpus Moment Retrieval
Mingjin Kuai · Qianyin Xiao · Juncheng Li · Jin Peng · Lizi Liao · Wei Ji
摘要
Video Corpus Moment Retrieval (VCMR) requires models to efficiently retrieve and precisely locate specific moments relevant to natural language queries within a massive, untrimmed video corpus. However, existing discriminative approaches typically rely on shallow visual-textual feature matching m...