Metraipolis › archive › dc_9243e8db305d

Beyond Relevance: Structured Semantic Supervision for Product Search with LLM-Augmented Annotations — arXiv:2609.23646

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.23646
License (as archived)
CC0-1.0
Archived
2026-10-09T06:50:38.133Z
SHA-256 of the archived text
0b6ef5c6a8e002958fc9785480bc8730d6725cb03b8941219d1534ee30c31dcf
Stable URL of this page
https://metraipolis.com/documento/dc_9243e8db305d

How to cite

Beyond Relevance: Structured Semantic Supervision for Product Search with LLM-Augmented Annotations — arXiv:2609.23646 — arXiv (en), https://arxiv.org/abs/2609.23646 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9243e8db305d (Metraipolis archive dc_9243e8db305d, sha256 0b6ef5c6a8e002958fc9785480bc8730d6725cb03b8941219d1534ee30c31dcf)
BibTeX
@misc{dc9243e8db305d,
  title = {Beyond Relevance: Structured Semantic Supervision for Product Search with LLM-Augmented Annotations — arXiv:2609.23646},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.23646},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9243e8db305d; sha256 0b6ef5c6a8e002958fc9785480bc8730d6725cb03b8941219d1534ee30c31dcf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23646. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

E-commerce search requires distinguishing products that are merely related to a query from those that directly satisfy the user's shopping intent. We augment query-product pairs with structured LLM-generated query and product attributes and human-validated relevance, explanations, and centrality judgments, and evaluate these signals using a simple dual-encoder retriever and MLP re-ranker. On an augmented subset of ESCI, a human-feature oracle reaches $0.9382$ nDCG@10, while a human-free trained $Q+P$ configuration reaches $0.9258$. Synthetic approximations of the human signals reach $0.9150$ overall but provide substantial gains for difficult, low-performing queries. Ablations show that most of the oracle improvement comes from post-edited explanations and annotator comments rather than the scalar centrality feature, suggesting that LLMs are most useful for exposing and approximating structured semantic supervision rather than replacing human judgment directly.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Information Retrieval (cs.IR) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.23646 — 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 2026-10-09T06:48:45.134Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).