Asking without Telling: Exploring Latent Ontologies in Contextual Representations
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10427025" target="_blank" >RIV/00216208:11320/20:10427025 - isvavai.cz</a>
Výsledek na webu
<a href="https://www.aclweb.org/anthology/2020.emnlp-main.552" target="_blank" >https://www.aclweb.org/anthology/2020.emnlp-main.552</a>
DOI - Digital Object Identifier
—
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Asking without Telling: Exploring Latent Ontologies in Contextual Representations
Popis výsledku v původním jazyce
The success of pretrained contextual encoders, such as ELMo and BERT, has brought a great deal of interest in what these models learn: do they, without explicit supervision, learn to encode meaningful notions of linguistic structure? If so, how is this structure encoded? To investigate this, we introduce latent subclass learning (LSL): a modification to classifier-based probing that induces a latent categorization (or ontology) of the probe's inputs. Without access to fine-grained gold labels, LSL extracts emergent structure from input representations in an interpretable and quantifiable form. In experiments, we find strong evidence of familiar categories, such as a notion of personhood in ELMo, as well as novel ontological distinctions, such as a preference for fine-grained semantic roles on core arguments. Our results provide unique new evidence of emergent structure in pretrained encoders, including departures from existing annotations which are inaccessible to earlier methods.
Název v anglickém jazyce
Asking without Telling: Exploring Latent Ontologies in Contextual Representations
Popis výsledku anglicky
The success of pretrained contextual encoders, such as ELMo and BERT, has brought a great deal of interest in what these models learn: do they, without explicit supervision, learn to encode meaningful notions of linguistic structure? If so, how is this structure encoded? To investigate this, we introduce latent subclass learning (LSL): a modification to classifier-based probing that induces a latent categorization (or ontology) of the probe's inputs. Without access to fine-grained gold labels, LSL extracts emergent structure from input representations in an interpretable and quantifiable form. In experiments, we find strong evidence of familiar categories, such as a notion of personhood in ELMo, as well as novel ontological distinctions, such as a preference for fine-grained semantic roles on core arguments. Our results provide unique new evidence of emergent structure in pretrained encoders, including departures from existing annotations which are inaccessible to earlier methods.
Klasifikace
Druh
O - Ostatní výsledky
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Návaznosti výsledku
Projekt
—
Návaznosti
—
Ostatní
Rok uplatnění
2020
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů