Metraipolis › archive › dc_f177f19d3bfe
SA-RSQ: A Versatile Sparse Representation Framework for Multi-modal Recommender Systems — arXiv:2608.22979
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.22979
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T07:41:45.067Z
- SHA-256 of the archived text
096529fb285a8a7e7894529440344360e5251124105a4470da642dc1a2b1ab34- Stable URL of this page
https://metraipolis.com/documento/dc_f177f19d3bfe
How to cite
SA-RSQ: A Versatile Sparse Representation Framework for Multi-modal Recommender Systems — arXiv:2608.22979 — arXiv (en), https://arxiv.org/abs/2608.22979 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f177f19d3bfe (Metraipolis archive dc_f177f19d3bfe, sha256 096529fb285a8a7e7894529440344360e5251124105a4470da642dc1a2b1ab34)
BibTeX
@misc{dcf177f19d3bfe,
title = {SA-RSQ: A Versatile Sparse Representation Framework for Multi-modal Recommender Systems — arXiv:2608.22979},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.22979},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f177f19d3bfe; sha256 096529fb285a8a7e7894529440344360e5251124105a4470da642dc1a2b1ab34},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22979. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Deploying high-dimensional multimodal features in industrial recommender systems incurs substantial storage and latency overhead. Hard quantization is compact but introduces boundary distortion, whereas dense soft quantization couples representation quality to the limited storage budget. We propose Sparse Activation-based Residual Soft Quantization (SA-RSQ), which uses Top-K sparse routing and softmax weights to store compact (Index, Probability) tuples. The stored tuples decouple per-item storage from codebook dimensionality; for a fixed selected support, gradients propagate through the routing weights and weighted reconstruction without relying on a straight-through estimator. Experiments on a proprietary food-delivery advertising dataset show favorable reconstruction-performance and CTR trade-offs across storage budgets of 8-48 bytes per item. A preliminary Next-Distribution Prediction study and a one-week online A/B test further demonstrate the practical potential of SA-RSQ, with relative lifts of +2.51% in CTR and +3.66% in CPM. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.22979 — 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 2026-10-18T07:37:52.757Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).