Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2407.14039 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916330254368768 |
|---|---|
| author | Saligram, Pradyumna Lanpouthakoun, Andrew |
| author_facet | Saligram, Pradyumna Lanpouthakoun, Andrew |
| contents | We explore advanced fine-tuning techniques to boost BERT's performance in sentiment analysis, paraphrase detection, and semantic textual similarity. Our approach leverages SMART regularization to combat overfitting, improves hyperparameter choices, employs a cross-embedding Siamese architecture for improved sentence embeddings, and introduces innovative early exiting methods. Our fine-tuning findings currently reveal substantial improvements in model efficiency and effectiveness when combining multiple fine-tuning architectures, achieving a state-of-the-art performance score of on the test set, surpassing current benchmarks and highlighting BERT's adaptability in multifaceted linguistic tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_14039 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | BERTer: The Efficient One Saligram, Pradyumna Lanpouthakoun, Andrew Computation and Language Machine Learning We explore advanced fine-tuning techniques to boost BERT's performance in sentiment analysis, paraphrase detection, and semantic textual similarity. Our approach leverages SMART regularization to combat overfitting, improves hyperparameter choices, employs a cross-embedding Siamese architecture for improved sentence embeddings, and introduces innovative early exiting methods. Our fine-tuning findings currently reveal substantial improvements in model efficiency and effectiveness when combining multiple fine-tuning architectures, achieving a state-of-the-art performance score of on the test set, surpassing current benchmarks and highlighting BERT's adaptability in multifaceted linguistic tasks. |
| title | BERTer: The Efficient One |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2407.14039 |