Abstract
The increasingly widespread adoption of large language models has highlighted the need for improving their explainability. We present *context length probing*, a novel explanation technique for causal language models, based on tracking the predictions of a model as a function of the length of available context, and allowing to assign *differential importance scores* to different contexts. The technique is model-agnostic and does not rely on access to model internals beyond computing token-level probabilities. We apply context length probing to large pre-trained language models and offer some initial analyses and insights, including the potential for studying long-range dependencies. The [source code](https://github.com/cifkao/context-probing/) and an [interactive demo](https://cifkao.github.io/context-probing/) of the method are available.- Anthology ID:
- 2023.acl-short.92
- Volume:
- Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
- Month:
- July
- Year:
- 2023
- Address:
- Toronto, Canada
- Editors:
- Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1067–1079
- Language:
- URL:
- https://aclanthology.org/2023.acl-short.92
- DOI:
- 10.18653/v1/2023.acl-short.92
- Cite (ACL):
- Ondřej Cífka and Antoine Liutkus. 2023. Black-box language model explanation by context length probing. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 1067–1079, Toronto, Canada. Association for Computational Linguistics.
- Cite (Informal):
- Black-box language model explanation by context length probing (Cífka & Liutkus, ACL 2023)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2023.acl-short.92.pdf