rgalhama / segmentation_as_unsup_parsing

Code for ACL 2022 paper.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About

Code for ACL-2022 paper:

  • Alhama, Raquel G. (2022). Segmentation as Unsupervised Constituency Parsing. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 4103--4112.

Howto

The stimuli is provided in data/. It can be regenerated with the stimuliPD.py script.

Run Bash scripts (/src/scripts/*.sh) for training, test and evaluation.

Scripts in src/analyses can be used to reproduce figures and analyses reported in the ACL2022 paper.

All the reported simulations are based on my forked version of DIORA (Drozdov et al., 2019), which removes dependency on pre-trained embeddings.

Whyto

goto #About

About

Code for ACL 2022 paper.

License:BSD 2-Clause "Simplified" License


Languages

Language:Python 84.8%Language:Shell 15.2%