Baked-in State Probing

Shubham Toshniwal, Sam Wiseman, Karen Livescu, Kevin Gimpel


Abstract
Neural language models have been analyzed for their linguistic and extra-linguistic knowledge via probing. Of particular interest has been the following question: how much can a language model trained only on form learn about meaning? Recent work has demonstrated via probing classifiers that in the setting of simple procedural text, where by “meaning” we mean the underlying world state, language models have a non-trivial performance on world state tracking. However, our proposed evaluation based on model predictions shows differing results, suggesting that these models are either not capturing the world state or not using it. How do these results change if the model has access to the world state? We explore this alternate setting with access to the underlying world state only during training and investigate ways of “baking in” the state knowledge along with the primary task of language modeling. Our proposed approaches allow for state probing during inference simply via text prompts, avoiding any probing classifier machinery. In terms of performance, we show that baking in the state knowledge during training leads to significant improvements in state tracking performance and text generation quality,
Anthology ID:
2022.findings-emnlp.397
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5430–5435
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.397
DOI:
10.18653/v1/2022.findings-emnlp.397
Bibkey:
Cite (ACL):
Shubham Toshniwal, Sam Wiseman, Karen Livescu, and Kevin Gimpel. 2022. Baked-in State Probing. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 5430–5435, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Baked-in State Probing (Toshniwal et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.397.pdf
Video:
 https://aclanthology.org/2022.findings-emnlp.397.mp4