Character-Level Chinese Backpack Language Models

Hao Sun, John Hewitt


Abstract
The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. However, Backpacks’ reliance on token-defined meaning raises questions as to their potential for languages other than English, a language for which subword tokenization provides a reasonable approximation for lexical items. In this work, we train, evaluate, interpret, and control Backpack language models in character-tokenized Chinese, in which words are often composed of many characters. We find that our (134M parameter) Chinese Backpack language model performs comparably to a (104M parameter) Transformer, and learns rich character-level meanings that log-additively compose to form word meanings. In SimLex-style lexical semantic evaluations, simple averages of Backpack character senses outperform input embeddings from a Transformer. We find that complex multi-character meanings are often formed by using the same per-character sense weights consistently across context. Exploring interpretability-through control, we show that we can localize a source of gender bias in our Backpacks to specific character senses and intervene to reduce the bias.
Anthology ID:
2023.blackboxnlp-1.8
Volume:
Proceedings of the 6th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
Month:
December
Year:
2023
Address:
Singapore
Editors:
Yonatan Belinkov, Sophie Hao, Jaap Jumelet, Najoung Kim, Arya McCarthy, Hosein Mohebbi
Venues:
BlackboxNLP | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
106–119
Language:
URL:
https://aclanthology.org/2023.blackboxnlp-1.8
DOI:
10.18653/v1/2023.blackboxnlp-1.8
Bibkey:
Cite (ACL):
Hao Sun and John Hewitt. 2023. Character-Level Chinese Backpack Language Models. In Proceedings of the 6th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 106–119, Singapore. Association for Computational Linguistics.
Cite (Informal):
Character-Level Chinese Backpack Language Models (Sun & Hewitt, BlackboxNLP-WS 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/naacl24-info/2023.blackboxnlp-1.8.pdf