Metraipolis › archive › dc_6f1ca92decbc

Scaling E-Commerce Attribute Extraction with Parallel Decoding — arXiv:2609.09716

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.09716
License (as archived): CC0-1.0
Archived: 2026-10-16T07:51:17.922Z
SHA-256 of the archived text: a89cad36a7db5c92640ab16a4d793fad0eef67a49f9347314882f41fb9b6950b
Stable URL of this page: https://metraipolis.com/documento/dc_6f1ca92decbc

How to cite

Scaling E-Commerce Attribute Extraction with Parallel Decoding — arXiv:2609.09716 — arXiv (en), https://arxiv.org/abs/2609.09716 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6f1ca92decbc (Metraipolis archive dc_6f1ca92decbc, sha256 a89cad36a7db5c92640ab16a4d793fad0eef67a49f9347314882f41fb9b6950b)

BibTeX

@misc{dc6f1ca92decbc,
  title = {Scaling E-Commerce Attribute Extraction with Parallel Decoding — arXiv:2609.09716},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09716},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6f1ca92decbc; sha256 a89cad36a7db5c92640ab16a4d793fad0eef67a49f9347314882f41fb9b6950b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09716. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Customers rely on specific product attributes to compare products and make purchasing decisions, but e-commerce catalogs are messy and unstructured, making it difficult to identify which attributes matter most and extract them at scale. Standard Attribute Value Extraction (AVE) systems treat all attributes equally, producing large, inconsistent attribute sets that do not reflect the factors consumers use to differentiate products. We introduce a two-stage LLM pipeline that first discovers a compact, ranked schema of purchase-discriminative attributes for each product category, then extracts their values from catalog text using a fine-tuned compact LLM (Qwen3-4B) with Hyper-Parallel Decoding (HPD). This pipeline achieves 85% extraction accuracy, on par with the foundational LLM it was distilled from, while reducing inference costs by 92% over foundational LLMs, enabling production-scale use for product discovery and catalog enrichment. The resulting category-level structured representations effectively constitute automatically constructed product knowledge bases, providing consistent, comparable attributes across varied product categories that can ground downstream knowledge-intensive applications.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-10 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.09716 — 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 2026-10-16T07:24:00.888Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).