{"ID":2851633,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.19410","arxiv_id":"2510.19410","title":"ToMMeR -- Efficient Entity Mention Detection from Large Language Models","abstract":"Identifying which text spans refer to entities - mention detection - is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a lightweight model (\u003c300K parameters) probing mention detection capabilities from early LLM layers. Across 13 NER benchmarks, ToMMeR achieves 93% recall zero-shot, with an estimated 90% precision under a human-calibrated LLM-judge protocol, showing that ToMMeR rarely produces spurious predictions despite high recall. Cross-model analysis reveals that diverse architectures (14M-15B parameters) converge on similar mention boundaries (DICE \u003e75%), confirming that mention detection emerges naturally from language modeling. When extended with span classification heads, ToMMeR achieves competitive NER performance (80-87% F1 on standard benchmarks). Our work provides evidence that structured entity representations exist in early transformer layers and can be efficiently recovered with minimal parameters.","short_abstract":"Identifying which text spans refer to entities - mention detection - is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a lightweight model (\u003c300K parameters) probing mention detection capabilities from early LLM layers. Across 13 NER benchmarks, ToMMeR achieves 93%...","url_abs":"https://arxiv.org/abs/2510.19410","url_pdf":"https://arxiv.org/pdf/2510.19410v2","authors":"[\"Victor Morand\",\"Nadi Tomeh\",\"Josiane Mothe\",\"Benjamin Piwowarski\"]","published":"2025-10-22T09:28:18Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Transformer\",\"Large Language Model\",\"Language Model\"]","has_code":false}