English

BERT-DRE: BERT with Deep Recursive Encoder for Natural Language Sentence Matching

Computation and Language 2021-11-05 v2

Abstract

This paper presents a deep neural architecture, for Natural Language Sentence Matching (NLSM) by adding a deep recursive encoder to BERT so called BERT with Deep Recursive Encoder (BERT-DRE). Our analysis of model behavior shows that BERT still does not capture the full complexity of text, so a deep recursive encoder is applied on top of BERT. Three Bi-LSTM layers with residual connection are used to design a recursive encoder and an attention module is used on top of this encoder. To obtain the final vector, a pooling layer consisting of average and maximum pooling is used. We experiment our model on four benchmarks, SNLI, FarsTail, MultiNLI, SciTail, and a novel Persian religious questions dataset. This paper focuses on improving the BERT results in the NLSM task. In this regard, comparisons between BERT-DRE and BERT are conducted, and it is shown that in all cases, BERT-DRE outperforms BERT. The BERT algorithm on the religious dataset achieved an accuracy of 89.70%, and BERT-DRE architectures improved to 90.29% using the same dataset.

Keywords

Cite

@article{arxiv.2111.02188,
  title  = {BERT-DRE: BERT with Deep Recursive Encoder for Natural Language Sentence Matching},
  author = {Ehsan Tavan and Ali Rahmati and Maryam Najafi and Saeed Bibak and Zahed Rahmati},
  journal= {arXiv preprint arXiv:2111.02188},
  year   = {2021}
}
R2 v1 2026-06-24T07:24:19.589Z