arXiv Open Access 2024

Tracking linguistic information in transformer-based sentence embeddings through targeted sparsification

Vivi Nastase Paola Merlo
Lihat Sumber

Abstrak

Analyses of transformer-based models have shown that they encode a variety of linguistic information from their textual input. While these analyses have shed a light on the relation between linguistic information on one side, and internal architecture and parameters on the other, a question remains unanswered: how is this linguistic information reflected in sentence embeddings? Using datasets consisting of sentences with known structure, we test to what degree information about chunks (in particular noun, verb or prepositional phrases), such as grammatical number, or semantic role, can be localized in sentence embeddings. Our results show that such information is not distributed over the entire sentence embedding, but rather it is encoded in specific regions. Understanding how the information from an input text is compressed into sentence embeddings helps understand current transformer models and help build future explainable neural models.

Topik & Kata Kunci

Penulis (2)

V

Vivi Nastase

P

Paola Merlo

Format Sitasi

Nastase, V., Merlo, P. (2024). Tracking linguistic information in transformer-based sentence embeddings through targeted sparsification. https://arxiv.org/abs/2407.18119

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2024
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓