An Evaluation of Language-Agnostic Inner-Attention-Based Representations in Machine Translation

Tutkimustuotos: Artikkeli kirjassa/raportissa/konferenssijulkaisussaKonferenssiartikkeliTieteellinenvertaisarvioitu

Abstrakti

In this paper, we explore a multilingual translation model with a cross-lingually shared layer that can be used as fixed-size sentence representation in different downstream tasks. We systematically study the impact of the size of the shared layer and the effect of including additional languages in the model. In contrast to related previous work, we demonstrate that the performance in translation does correlate with trainable downstream tasks. In particular, we show that larger intermediate layers not only improve translation quality, especially for long sentences, but also push the accuracy of trainable classification tasks. On the other hand, shorter representations lead to increased compression that is beneficial in non-trainable similarity tasks. We hypothesize that the training procedure on the downstream task enables the model to identify the encoded information that is useful for the specific task whereas non-trainable benchmarks can be confused by other types of information also encoded in the representation of a sentence.
Alkuperäiskielienglanti
OtsikkoThe 4th Workshop on Representation Learning for NLP (RepL4NLP-2019) : Proceedings of the Workshop
ToimittajatIsabelle Augenstein, Spandana Gella, Sebastian Ruder, Katharina Kann, Burcu Can, Johannes Welbl, Alexis Conneau, Xiang Ren, Marek Rei
Sivumäärä6
JulkaisupaikkaStroudsburg
KustantajaThe Association for Computational Linguistics
Julkaisupäivä1 elok. 2019
Sivut27-32
ISBN (elektroninen)978-1-950737-35-2
TilaJulkaistu - 1 elok. 2019
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisuussa
TapahtumaWorkshop on Representation Learning for NLP - Florence, Italia
Kesto: 2 elok. 20192 elok. 2019
Konferenssinumero: 4

Tieteenalat

  • 6121 Kielitieteet
  • 113 Tietojenkäsittely- ja informaatiotieteet

Siteeraa tätä