From Robustness to Improved Generalization and Calibration in Pre-trained Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914736404168704 |
|---|---|
| author | Jukić, Josip Šnajder, Jan |
| author_facet | Jukić, Josip Šnajder, Jan |
| contents | Enhancing generalization and uncertainty quantification in pre-trained language models (PLMs) is crucial for their effectiveness and reliability. Building on machine learning research that established the importance of robustness for improving generalization, we investigate the role of representation smoothness, achieved via Jacobian and Hessian regularization, in enhancing PLM performance. Although such regularization methods have proven effective in computer vision, their application in natural language processing (NLP), where PLM inputs are derived from a discrete domain, poses unique challenges. We introduce a novel two-phase regularization approach, JacHess, which minimizes the norms of the Jacobian and Hessian matrices within PLM intermediate representations relative to their inputs. Our evaluation using the GLUE benchmark demonstrates that JacHess significantly improves in-domain generalization and calibration in PLMs, outperforming unregularized fine-tuning and other similar regularization methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_00758 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | From Robustness to Improved Generalization and Calibration in Pre-trained Language Models Jukić, Josip Šnajder, Jan Computation and Language Enhancing generalization and uncertainty quantification in pre-trained language models (PLMs) is crucial for their effectiveness and reliability. Building on machine learning research that established the importance of robustness for improving generalization, we investigate the role of representation smoothness, achieved via Jacobian and Hessian regularization, in enhancing PLM performance. Although such regularization methods have proven effective in computer vision, their application in natural language processing (NLP), where PLM inputs are derived from a discrete domain, poses unique challenges. We introduce a novel two-phase regularization approach, JacHess, which minimizes the norms of the Jacobian and Hessian matrices within PLM intermediate representations relative to their inputs. Our evaluation using the GLUE benchmark demonstrates that JacHess significantly improves in-domain generalization and calibration in PLMs, outperforming unregularized fine-tuning and other similar regularization methods. |
| title | From Robustness to Improved Generalization and Calibration in Pre-trained Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2404.00758 |