Metraipolis › archive › dc_01deb1b28bd5

MELON: A Large-Scale Dataset for Multi-Event Text-to-Long-Video Retrieval — arXiv:2609.01654

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01654
License (as archived)
CC0-1.0
Archived
2026-10-17T05:46:45.039Z
SHA-256 of the archived text
45155e1b83fae7a40419724b1f3d193b7736bee3765a5b07520f81c690580058
Stable URL of this page
https://metraipolis.com/documento/dc_01deb1b28bd5

How to cite

MELON: A Large-Scale Dataset for Multi-Event Text-to-Long-Video Retrieval — arXiv:2609.01654 — arXiv (en), https://arxiv.org/abs/2609.01654 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_01deb1b28bd5 (Metraipolis archive dc_01deb1b28bd5, sha256 45155e1b83fae7a40419724b1f3d193b7736bee3765a5b07520f81c690580058)
BibTeX
@misc{dc01deb1b28bd5,
  title = {MELON: A Large-Scale Dataset for Multi-Event Text-to-Long-Video Retrieval — arXiv:2609.01654},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01654},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_01deb1b28bd5; sha256 45155e1b83fae7a40419724b1f3d193b7736bee3765a5b07520f81c690580058},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01654. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Existing text-video retrieval datasets primarily consist of short-form clips containing a single dominant event. While suitable for measuring basic vision-language alignment, they are limited in capturing real-world retrieval scenarios, where long-form videos naturally contain multiple semantically distinct events and a single text query may correspond to several non-contiguous temporal segments. To bridge this gap, we introduce MELON, the first large-scale dataset designed to extend text-video retrieval to long-form videos featuring complex, multi-event structures. MELON explicitly annotates multiple event intervals per video along with their corresponding textual descriptions, enabling both training and evaluation of multi-event understanding in long, untrimmed videos. In addition, we propose a multi-event aware loss that encourages models to differentiate between full-event and partial-event matches, yielding substantial improvements in retrieval accuracy. Together, the MELON dataset and our proposed loss establish a robust foundation for expanding text-to-video retrieval to complex long-form scenarios and provide a more realistic evaluation setting for future research in the field.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-03 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2609.01654 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T05:23:20.828Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).