Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2401.08511 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916093453402112 |
|---|---|
| author | Kaneko, Masahiro Bollegala, Danushka Baldwin, Timothy |
| author_facet | Kaneko, Masahiro Bollegala, Danushka Baldwin, Timothy |
| contents | The output tendencies of Pre-trained Language Models (PLM) vary markedly before and after Fine-Tuning (FT) due to the updates to the model parameters. These divergences in output tendencies result in a gap in the social biases of PLMs. For example, there exits a low correlation between intrinsic bias scores of a PLM and its extrinsic bias scores under FT-based debiasing methods. Additionally, applying FT-based debiasing methods to a PLM leads to a decline in performance in downstream tasks. On the other hand, PLMs trained on large datasets can learn without parameter updates via In-Context Learning (ICL) using prompts. ICL induces smaller changes to PLMs compared to FT-based debiasing methods. Therefore, we hypothesize that the gap observed in pre-trained and FT models does not hold true for debiasing methods that use ICL. In this study, we demonstrate that ICL-based debiasing methods show a higher correlation between intrinsic and extrinsic bias scores compared to FT-based methods. Moreover, the performance degradation due to debiasing is also lower in the ICL case compared to that in the FT case. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_08511 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | The Gaps between Pre-train and Downstream Settings in Bias Evaluation and Debiasing Kaneko, Masahiro Bollegala, Danushka Baldwin, Timothy Computation and Language The output tendencies of Pre-trained Language Models (PLM) vary markedly before and after Fine-Tuning (FT) due to the updates to the model parameters. These divergences in output tendencies result in a gap in the social biases of PLMs. For example, there exits a low correlation between intrinsic bias scores of a PLM and its extrinsic bias scores under FT-based debiasing methods. Additionally, applying FT-based debiasing methods to a PLM leads to a decline in performance in downstream tasks. On the other hand, PLMs trained on large datasets can learn without parameter updates via In-Context Learning (ICL) using prompts. ICL induces smaller changes to PLMs compared to FT-based debiasing methods. Therefore, we hypothesize that the gap observed in pre-trained and FT models does not hold true for debiasing methods that use ICL. In this study, we demonstrate that ICL-based debiasing methods show a higher correlation between intrinsic and extrinsic bias scores compared to FT-based methods. Moreover, the performance degradation due to debiasing is also lower in the ICL case compared to that in the FT case. |
| title | The Gaps between Pre-train and Downstream Settings in Bias Evaluation and Debiasing |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2401.08511 |