Semiparametric Token-Sequence Co-Supervision

Hyunji Lee, Doyoung Kim, Jihoon Jun, Se June Joo, Joel Jang, Kyoung-Woon On, Minjoon Seo


Abstract
In this work, we introduce a semiparametric token-sequence co-supervision training method. It trains a language model by simultaneously leveraging supervision from the traditional next token prediction loss which is calculated over the parametric token embedding space and the next sequence prediction loss which is calculated over the nonparametric sequence embedding space. The nonparametric sequence embedding space is constructed by a separate language model tasked to condense an input text into a single representative embedding. Our experiments demonstrate that a model trained via both supervisions consistently surpasses models trained via each supervision independently. Analysis suggests that this co-supervision encourages a broader generalization capability across the model. Especially, the robustness of parametric token space which is established during the pretraining step tends to effectively enhance the stability of nonparametric sequence embedding space, a new space established by another language model.
Anthology ID:
2024.acl-long.213
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3864–3882
Language:
URL:
https://aclanthology.org/2024.acl-long.213
DOI:
10.18653/v1/2024.acl-long.213
Bibkey:
Cite (ACL):
Hyunji Lee, Doyoung Kim, Jihoon Jun, Se June Joo, Joel Jang, Kyoung-Woon On, and Minjoon Seo. 2024. Semiparametric Token-Sequence Co-Supervision. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3864–3882, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Semiparametric Token-Sequence Co-Supervision (Lee et al., ACL 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/autopr/2024.acl-long.213.pdf