Abstract
Next-word predictions from autoregressive neural language models show remarkable sensitivity to syntax. This work evaluates the extent to which this behavior arises as a result of a learned ability to maintain implicit representations of incremental syntactic structures. We extend work in syntactic probing to the incremental setting and present several probes for extracting incomplete syntactic structure (operationalized through parse states from a stack-based parser) from autoregressive language models. We find that our probes can be used to predict model preferences on ambiguous sentence prefixes and causally intervene on model representations and steer model behavior. This suggests implicit incremental syntactic inferences underlie next-word predictions in autoregressive neural language models.- Anthology ID:
- 2022.findings-emnlp.203
- 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:
- 2801–2813
- Language:
- URL:
- https://aclanthology.org/2022.findings-emnlp.203
- DOI:
- 10.18653/v1/2022.findings-emnlp.203
- Cite (ACL):
- Tiwalayo Eisape, Vineet Gangireddy, Roger Levy, and Yoon Kim. 2022. Probing for Incremental Parse States in Autoregressive Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 2801–2813, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- Probing for Incremental Parse States in Autoregressive Language Models (Eisape et al., Findings 2022)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2022.findings-emnlp.203.pdf