Metraipolis › archive › dc_30bc313a9e17

Select, Don't Train: The Benefits of Modular Entity Disambiguation with LLM-Based Selection — arXiv:2608.27470

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.27470
License (as archived)
CC0-1.0
Archived
2026-10-17T10:47:45.380Z
SHA-256 of the archived text
8c1fc62c40f71b4c4e91288d7d362e6a2faae14e23bf7069ff9f28c577e72e22
Stable URL of this page
https://metraipolis.com/documento/dc_30bc313a9e17

How to cite

Select, Don't Train: The Benefits of Modular Entity Disambiguation with LLM-Based Selection — arXiv:2608.27470 — arXiv (en), https://arxiv.org/abs/2608.27470 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_30bc313a9e17 (Metraipolis archive dc_30bc313a9e17, sha256 8c1fc62c40f71b4c4e91288d7d362e6a2faae14e23bf7069ff9f28c577e72e22)
BibTeX
@misc{dc30bc313a9e17,
  title = {Select, Don't Train: The Benefits of Modular Entity Disambiguation with LLM-Based Selection — arXiv:2608.27470},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27470},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_30bc313a9e17; sha256 8c1fc62c40f71b4c4e91288d7d362e6a2faae14e23bf7069ff9f28c577e72e22},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27470. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Entity Disambiguation (ED) is a key task for constructing and using knowledge graphs. State-of-the-art neural approaches commonly model ED as a single task, although it consists of two distinct subproblems: retrieving candidate entities and selecting the correct one given context. Dual-encoder models optimize for both within a shared embedding space, forcing representations to balance high-recall retrieval with fine-grained selection, and they require trained retrievers, which are costly to maintain as knowledge graphs change. While recent work has begun to combine retrievers with LLM-based selectors, the interplay between the two stages has not been studied systematically. In this paper, we present a systematic comparison of retrieval strategies for candidate generation under a shared LLM-based selection stage, combining sparse retrieval (BM25), Web KB search, and a state-of-the-art trained dense retriever with several open- and closed-source LLMs. We show that, once selection is delegated to a capable LLM, training the retriever provides only modest additional value: a fully training-free BM25 retriever paired with an LLM selector reaches a new state of the art on the ZELDA benchmark, raising inKB micro-F1 from 82.3 to 86.3 (+4); pairing the same LLM with a trained dense retriever reaches 88.5. Decoupling retrieval from selection also exposes a limitation of current ED systems: when the correct entity is missing from retrieved candidates, they are forced to predict an incorrect entity. In contrast, our framework allows for abstention when retrieval failure is detected. In an evaluation setting that rewards correct abstentions, the training-free BM25 + LLM pipeline reaches 90.7 F1.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Databases (cs.DB).
Pagina del lavoro: https://arxiv.org/abs/2608.27470 — 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 2026-10-17T10:33:42.442Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).