arXiv Open Access 2023

Improving Grammar-based Sequence-to-Sequence Modeling with Decomposition and Constraints

Chao Lou Kewei Tu

Lihat Sumber

Abstrak

Neural QCFG is a grammar-based sequence-tosequence (seq2seq) model with strong inductive biases on hierarchical structures. It excels in interpretability and generalization but suffers from expensive inference. In this paper, we study two low-rank variants of Neural QCFG for faster inference with different trade-offs between efficiency and expressiveness. Furthermore, utilizing the symbolic interface provided by the grammar, we introduce two soft constraints over tree hierarchy and source coverage. We experiment with various datasets and find that our models outperform vanilla Neural QCFG in most settings.

Topik & Kata Kunci

cs.CL

Penulis (2)

Chao Lou

Kewei Tu

Format Sitasi

APA MLA BibTeX

Lou, C., Tu, K. (2023). Improving Grammar-based Sequence-to-Sequence Modeling with Decomposition and Constraints. https://arxiv.org/abs/2306.02671

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2023
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓