Life after BERT: What do Other Muppets Understand about Language?
V. Lialin, K. Zhao, N. Shivagunde, A. Rumshisky., ACL 2022
Paper linkExisting pre-trained transformer analysis works usually focus only on one or two model families at a time, overlooking the variability of the architecture and pre-training objectives. In our work, we utilize the oLMpics bench- mark and psycholinguistic probing datasets for a diverse set of 29 models including T5, BART, and ALBERT. Additionally, we adapt the oLMpics zero-shot setup for autoregres- sive models and evaluate GPT networks of different sizes. Our findings show that none of these models can resolve compositional questions in a zero-shot fashion, suggesting that this skill is not learnable using existing pre-training objectives. Furthermore, we find that global model decisions such as architecture, directionality, size of the dataset, and pre-training objective are not predictive of a model’s linguistic capabilities.
Citation:
@inproceedings{lialin-etal-2022-life,
title = "Life after {BERT}: What do Other Muppets Understand about Language?",
author = "Lialin, Vladislav and
Zhao, Kevin and
Shivagunde, Namrata and
Rumshisky, Anna",
booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = may,
year = "2022",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.acl-long.227",
doi = "10.18653/v1/2022.acl-long.227",
pages = "3180--3193",
}