@inproceedings{shachar-etal-2025-ner,
    title = "{NER} Retriever: Zero-Shot Named Entity Retrieval with Type-Aware Embeddings",
    author = "Shachar, Or  and
      Katz, Uri  and
      Goldberg, Yoav  and
      Glickman, Oren",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.597/",
    doi = "10.18653/v1/2025.findings-emnlp.597",
    pages = "11175--11186",
    ISBN = "979-8-89176-335-7",
    abstract = "We present NER Retriever, a zero-shot retrieval framework for ad-hoc Named Entity Recognition (NER), where a user-defined type description is used to retrieve documents mentioning entities of that type. Instead of relying on fixed schemas or fine-tuned models, our method builds on pretrained language models (LLMs) to embed both entity mentions and type descriptions into a shared semantic space. We show that internal representations{---}specifically, the value vectors from mid-layer transformer blocks{---}encode fine-grained type information more effectively than commonly used top-layer embeddings. To refine these representations, we train a lightweight contrastive projection network that aligns type-compatible entities while separating unrelated types. The resulting entity embeddings are compact, type-aware, and well-suited for nearest-neighbor search. Evaluated on three benchmarks, NER Retriever significantly outperforms both lexical (BM25) and dense (sentence-level) retrieval baselines, particularly in low-context settings. Our findings provide empirical support for representation selection within LLMs and demonstrate a practical solution for scalable, schema-free entity retrieval."
}