Metraipolis › archive › dc_9d45161dc49d

arXiv 2610.28435: Multimodal Reranking Is a System One Decision

Source
Metraipolis archive
Original
https://arxiv.org/abs/2610.28435
License (as archived)
CC-BY-4.0
Archived
2026-10-21T06:06:24.963Z
SHA-256 of the archived text
4aefa2a7e3aa402931432e1606d873d4a4dca16f4a50c2356ad362f54d5995e6
Stable URL of this page
https://metraipolis.com/documento/dc_9d45161dc49d

How to cite

arXiv 2610.28435: Multimodal Reranking Is a System One Decision — Metraipolis archive, https://arxiv.org/abs/2610.28435 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_9d45161dc49d (Metraipolis archive dc_9d45161dc49d, sha256 4aefa2a7e3aa402931432e1606d873d4a4dca16f4a50c2356ad362f54d5995e6)
BibTeX
@misc{dc9d45161dc49d,
  title = {arXiv 2610.28435: Multimodal Reranking Is a System One Decision},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.28435},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_9d45161dc49d; sha256 4aefa2a7e3aa402931432e1606d873d4a4dca16f4a50c2356ad362f54d5995e6},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.28435. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal rerankers are usually dedicated models: a vision-language model (VLM) is trained on retrieval data to score the candidates of a query. Scoring options is not specific to reranking. System One decision models such as Jev map a state and a typed question to one probability per option without generating text, and general models can make such decisions with little or no training. Reranking is one such decision, with the query and its candidates as the state and the candidates as the options. We therefore argue that a general VLM can serve as a multimodal reranker through System One decisions, and that decision tuning can improve its reranking. We ask a VLM one choice question over the top 16 candidates of a first-stage retriever and rank the candidates by the probabilities of their labels, read from one forward pass. On MMEB-V2 image retrieval, this choice scores 5.4 to 7.3 hit@1 above a yes/no question per candidate, and the yes/no question falls below the first stage. With the choice, zero-shot Qwen3.8-27B beats the supervised Qwen3-VL-Reranker-2B on all three benchmarks and comes within 1.6 points of Reranker-8B on each. It matches Reranker-8B on MMEB-V2 visual documents (84.79 vs. 84.78 nDCG@5), at 1.1 to 1.5 times its latency. JEV-27B-VL, the same backbone tuned on text decisions only, improves page reranking by 0.6 to 1.1 points, with gains on most tasks (sign test, p <= 0.017), and leaves natural images unchanged. The zero-shot model is larger than the supervised rerankers and of a newer generation, so these comparisons are confounded; we discuss this and other alternative views, and the experiments that would test the position.

Articolo: «Multimodal Reranking Is a System One Decision» — Liu, Zhuchenyang; Zhang, Yao; Xiao, Yu (2026-10-20), arXiv:2610.28435, categoria cs.IR (ricerca dell'informazione).
Fonte: https://arxiv.org/abs/2610.28435
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-21T06:06:16Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.