{"ID":6023440,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T07:58:41.377760688Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.05971","arxiv_id":"2607.05971","title":"Multimodal Video-to-Music Recommendation via Semantic Retrieval and Temporal Reranking","abstract":"We present VTMR, a two-stage framework for Video-To-Music Recommendation. In Stage~1, VTMR aligns comprehensive video and music signals in a joint audio-visual-text representation space and efficiently retrieves semantically compatible candidates using coarse global embeddings. In Stage~2, it reranks the retrieved candidates by attending to the temporal sequences of both video and music, thereby capturing fine-grained temporal correspondence. Evaluated on the video-to-music recommendation task, the multimodal retrieval stage improves R@10 from 14.2 to 15.9 and Median Rank from 75 to 58 over the strongest baseline; the temporal reranker further boosts R@10 to 18.3 and Median Rank to 46, demonstrating complementary gains from richer query encoding and temporal alignment. A human preference study confirms that VTMR is on par with a commercial baseline in overall preference, while outperforming a generative baseline in music quality.","short_abstract":"We present VTMR, a two-stage framework for Video-To-Music Recommendation. In Stage~1, VTMR aligns comprehensive video and music signals in a joint audio-visual-text representation space and efficiently retrieves semantically compatible candidates using coarse global embeddings. In Stage~2, it reranks the retrieved cand...","url_abs":"https://arxiv.org/abs/2607.05971","url_pdf":"https://arxiv.org/pdf/2607.05971v1","authors":"[\"Seungheon Doh\",\"Minhee Lee\",\"Sangmoon Lee\",\"Ben Sangbae Chon\",\"Juhan Nam\"]","published":"2026-07-07T08:04:56Z","proceeding":"cs.MM","tasks":"[\"cs.MM\",\"cs.SD\"]","methods":"[]","has_code":false}