Metraipolis › archive › dc_a564b6f656aa
RandSlot: Learning Compact Visual Document Representations with Random Soft Tokens — arXiv:2609.32699
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32699
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:49:47.508Z
- SHA-256 of the archived text
0fdcf6f4c2cf74eece32123c6debd96d5a4ee543c275a34847e0a7e02993b6a5- Stable URL of this page
https://metraipolis.com/documento/dc_a564b6f656aa
How to cite
RandSlot: Learning Compact Visual Document Representations with Random Soft Tokens — arXiv:2609.32699 — arXiv (en), https://arxiv.org/abs/2609.32699 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a564b6f656aa (Metraipolis archive dc_a564b6f656aa, sha256 0fdcf6f4c2cf74eece32123c6debd96d5a4ee543c275a34847e0a7e02993b6a5)
BibTeX
@misc{dca564b6f656aa,
title = {RandSlot: Learning Compact Visual Document Representations with Random Soft Tokens — arXiv:2609.32699},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32699},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a564b6f656aa; sha256 0fdcf6f4c2cf74eece32123c6debd96d5a4ee543c275a34847e0a7e02993b6a5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32699. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Visual document retrieval requires expressive representations to match queries with evidence distributed across text, tables, and page layouts. Multi-vector representations capture fine-grained information, but storing and comparing many vectors introduces substantial retrieval costs. In this paper, we introduce RandSlot, a simple approach to learn compact visual document representations with random soft tokens. During training, we append independently-sampled random unit vectors to query and document input sequences and resample them at every use, without introducing learnable soft-token parameters. The encoder contextualizes these auxiliary inputs with the original content to produce a small set of retrieval vectors. A standard late-interaction objective trains the encoder to extract relevant information under varying input conditions. Experiments with different backbone models show that RandSlot improves retrieval quality over alternative readout strategies under the same vector budget. Further analysis shows that these gains can persist when random soft tokens are replaced with zeros at inference, demonstrating that random inputs during training can improve compact retrieval representations even when inference no longer requires sampling. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.IR). Categorie arXiv: Information Retrieval (cs.IR). Pagina del lavoro: https://arxiv.org/abs/2609.32699 — 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 2026-10-07T23:26:57.182Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).