Metraipolis › archive › dc_2279a7fde028

Hybrid Generative-Tagging Decoding for Efficient Product Attribute-Value Extraction — arXiv:2610.21329

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.21329
License (as archived)
CC0-1.0
Archived
2026-10-16T09:56:01.007Z
SHA-256 of the archived text
cf30e5ba23fb6b1c22efc92edca937451a5f2190feb3cf0a0b84744f40cc4572
Stable URL of this page
https://metraipolis.com/documento/dc_2279a7fde028

How to cite

Hybrid Generative-Tagging Decoding for Efficient Product Attribute-Value Extraction — arXiv:2610.21329 — arXiv (en), https://arxiv.org/abs/2610.21329 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2279a7fde028 (Metraipolis archive dc_2279a7fde028, sha256 cf30e5ba23fb6b1c22efc92edca937451a5f2190feb3cf0a0b84744f40cc4572)
BibTeX
@misc{dc2279a7fde028,
  title = {Hybrid Generative-Tagging Decoding for Efficient Product Attribute-Value Extraction — arXiv:2610.21329},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.21329},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2279a7fde028; sha256 cf30e5ba23fb6b1c22efc92edca937451a5f2190feb3cf0a0b84744f40cc4572},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21329. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Attribute-value extraction is important for building high-quality product catalogs in e-commerce. Existing systems either rely on generative or tagging strategies for extracting the attribute names and their corresponding values. Tagging models directly label spans in the source, making them fast and grounded, but they cannot produce values that require reformatting or generate rare attribute names outside their label space. Generative models avoid these drawbacks by generating both attribute names and values token by token, but this auto-regressive nature makes them slow and limits attribute coverage. We introduce hybrid generative-tagging decoding, a novel model architecture that combines tagging source spans and open-ended generation. For frequent attributes whose values are copyable, a tagging head extracts values directly from the source. This lets the generative head focus on the remainder - rare attributes outside the tagger's label space and non-copyable values, including implicit or normalized values. Integrated into vLLM inference, the hybrid decoder raises attribute coverage by 10% and extracts 63% more correct attribute-value pairs at no precision cost, while reducing latency by 44%.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.21329 — 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 2026-10-16T08:44:47.639Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).