Logo Logo
Hilfe
Hilfe
Switch Language to English

Dufter, Philipp; Kassner, Nora und Schütze, Hinrich (Juni 2021): Static Embeddings as Efficient Knowledge Bases? 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Online, June 6-11, 2021. Toutanova, Kristina (Hrsg.): In: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Stroudsburg, PA: Association for Computational Linguistics. S. 2353-2363 [PDF, 340kB]

[thumbnail of 2021.naacl-main.186.pdf]
Vorschau
Download (340kB)

Abstract

Recent research investigates factual knowledge stored in large pretrained language models (PLMs). Instead of structural knowledge base (KB) queries, masked sentences such as “Paris is the capital of [MASK]” are used as probes. The good performance on this analysis task has been interpreted as PLMs becoming potential repositories of factual knowledge. In experiments across ten linguistically diverse languages, we study knowledge contained in static embeddings. We show that, when restricting the output space to a candidate set, simple nearest neighbor matching using static embeddings performs better than PLMs. E.g., static embeddings perform 1.6% points better than BERT while just using 0.3% of energy for training. One important factor in their good comparative performance is that static embeddings are standardly learned for a large vocabulary. In contrast, BERT exploits its more sophisticated, but expensive ability to compose meaningful representations from a much smaller subword vocabulary.

Dokument bearbeiten Dokument bearbeiten