Metraipolis › archive › dc_0799b66b050e

MARS: What Retrieval Signals Are Hidden in Multimodal Large Language Models for Text-Video Retrieval? — arXiv:2609.02565

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.02565
License (as archived)
CC0-1.0
Archived
2026-10-17T05:50:09.876Z
SHA-256 of the archived text
c6ba3ffb6f3c2f9e4766461e8ff7339fbd74e833daf9be887c11245e4bb7263c
Stable URL of this page
https://metraipolis.com/documento/dc_0799b66b050e

How to cite

MARS: What Retrieval Signals Are Hidden in Multimodal Large Language Models for Text-Video Retrieval? — arXiv:2609.02565 — arXiv (en), https://arxiv.org/abs/2609.02565 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0799b66b050e (Metraipolis archive dc_0799b66b050e, sha256 c6ba3ffb6f3c2f9e4766461e8ff7339fbd74e833daf9be887c11245e4bb7263c)
BibTeX
@misc{dc0799b66b050e,
  title = {MARS: What Retrieval Signals Are Hidden in Multimodal Large Language Models for Text-Video Retrieval? — arXiv:2609.02565},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.02565},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0799b66b050e; sha256 c6ba3ffb6f3c2f9e4766461e8ff7339fbd74e833daf9be887c11245e4bb7263c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02565. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Text-video retrieval requires representations that can distinguish videos with similar scenes, actions, and temporal patterns. Recent multimodal large language models have been adapted as embedding models, but they often represent each input using a single token from the final layer. This can compress diverse video-text cues into a single vector and limit fine-grained retrieval. To address this limitation, we propose MARS, a multi-layer and multi-slot embedding framework for text-video retrieval. MARS constructs multiple adaptive representation slots by combining hidden states from different decoder layers, compares corresponding text and video slots, and aggregates their similarities for retrieval. To better handle confusing candidates, we further introduce a hard-negative-aware slot specialization objective that encourages the slots to capture discriminative matching cues. Experiments on four text-video retrieval benchmarks show that MARS achieves state-of-the-art results in both direct similarity-based retrieval and reranking settings. Ablation studies and analyses demonstrate that multi-layer fusion, multiple slots, and hard-negative-aware slot specialization provide complementary gains. Code is available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-03 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.02565 — 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 2026-10-17T05:25:58.137Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).