Abstract
We investigate how different domains are encoded in modern neural network architectures. We analyze the relationship between natural language domains, model size, and the amount of training data used. The primary analysis tool we develop is based on subpopulation analysis with Singular Vector Canonical Correlation Analysis (SVCCA), which we apply to Transformer-based language models (LMs). We compare the latent representations of such a language model at its different layers from a pair of models: a model trained on multiple domains (an experimental model) and a model trained on a single domain (a control model). Through our method, we find that increasing the model capacity impacts how domain information is stored in upper and lower layers differently. In addition, we show that larger experimental models simultaneously embed domain-specific information as if they were conjoined control models. These findings are confirmed qualitatively, demonstrating the validity of our method.- Anthology ID:
- 2022.blackboxnlp-1.16
- Volume:
- Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates (Hybrid)
- Editors:
- Jasmijn Bastings, Yonatan Belinkov, Yanai Elazar, Dieuwke Hupkes, Naomi Saphra, Sarah Wiegreffe
- Venue:
- BlackboxNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 192–209
- Language:
- URL:
- https://aclanthology.org/2022.blackboxnlp-1.16
- DOI:
- 10.18653/v1/2022.blackboxnlp-1.16
- Cite (ACL):
- Zheng Zhao, Yftah Ziser, and Shay Cohen. 2022. Understanding Domain Learning in Language Models Through Subpopulation Analysis. In Proceedings of the Fifth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 192–209, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
- Cite (Informal):
- Understanding Domain Learning in Language Models Through Subpopulation Analysis (Zhao et al., BlackboxNLP 2022)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2022.blackboxnlp-1.16.pdf