Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2506.02701 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913949312614400 |
|---|---|
| author | Sakata, Masaki Heinzerling, Benjamin Yokoi, Sho Ito, Takumi Inui, Kentaro |
| author_facet | Sakata, Masaki Heinzerling, Benjamin Yokoi, Sho Ito, Takumi Inui, Kentaro |
| contents | We analyze the extent to which internal representations of language models (LMs) identify and distinguish mentions of named entities, focusing on the many-to-many correspondence between entities and their mentions. We first formulate two problems of entity mentions -- ambiguity and variability -- and propose a framework analogous to clustering quality metrics. Specifically, we quantify through cluster analysis of LM internal representations the extent to which mentions of the same entity cluster together and mentions of different entities remain separated. Our experiments examine five Transformer-based autoregressive models, showing that they effectively identify and distinguish entities with metrics analogous to precision and recall ranging from 0.66 to 0.9. Further analysis reveals that entity-related information is compactly represented in a low-dimensional linear subspace at early LM layers. Additionally, we clarify how the characteristics of entity representations influence word prediction performance. These findings are interpreted through the lens of isomorphism between LM representations and entity-centric knowledge structures in the real world, providing insights into how LMs internally organize and use entity information. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_02701 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On Entity Identification in Language Models Sakata, Masaki Heinzerling, Benjamin Yokoi, Sho Ito, Takumi Inui, Kentaro Computation and Language We analyze the extent to which internal representations of language models (LMs) identify and distinguish mentions of named entities, focusing on the many-to-many correspondence between entities and their mentions. We first formulate two problems of entity mentions -- ambiguity and variability -- and propose a framework analogous to clustering quality metrics. Specifically, we quantify through cluster analysis of LM internal representations the extent to which mentions of the same entity cluster together and mentions of different entities remain separated. Our experiments examine five Transformer-based autoregressive models, showing that they effectively identify and distinguish entities with metrics analogous to precision and recall ranging from 0.66 to 0.9. Further analysis reveals that entity-related information is compactly represented in a low-dimensional linear subspace at early LM layers. Additionally, we clarify how the characteristics of entity representations influence word prediction performance. These findings are interpreted through the lens of isomorphism between LM representations and entity-centric knowledge structures in the real world, providing insights into how LMs internally organize and use entity information. |
| title | On Entity Identification in Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2506.02701 |