Gaussian Stochastic Weight Averaging for Bayesian Low-Rank Adaptation of Large Language Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910536412692480 |
|---|---|
| author | Onal, Emre Flöge, Klemens Caldwell, Emma Sheverdin, Arsen Fortuin, Vincent |
| author_facet | Onal, Emre Flöge, Klemens Caldwell, Emma Sheverdin, Arsen Fortuin, Vincent |
| contents | Fine-tuned Large Language Models (LLMs) often suffer from overconfidence and poor calibration, particularly when fine-tuned on small datasets. To address these challenges, we propose a simple combination of Low-Rank Adaptation (LoRA) with Gaussian Stochastic Weight Averaging (SWAG), facilitating approximate Bayesian inference in LLMs. Through extensive testing across several Natural Language Processing (NLP) benchmarks, we demonstrate that our straightforward and computationally efficient approach improves model generalization and calibration competitively with comparable, more sophisticated methods for Bayesian inference in LLMs. We further show that our method exhibits greater robustness against distribution shift, as reflected in its improved performance on out-of-distribution tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_03425 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Gaussian Stochastic Weight Averaging for Bayesian Low-Rank Adaptation of Large Language Models Onal, Emre Flöge, Klemens Caldwell, Emma Sheverdin, Arsen Fortuin, Vincent Computation and Language Fine-tuned Large Language Models (LLMs) often suffer from overconfidence and poor calibration, particularly when fine-tuned on small datasets. To address these challenges, we propose a simple combination of Low-Rank Adaptation (LoRA) with Gaussian Stochastic Weight Averaging (SWAG), facilitating approximate Bayesian inference in LLMs. Through extensive testing across several Natural Language Processing (NLP) benchmarks, we demonstrate that our straightforward and computationally efficient approach improves model generalization and calibration competitively with comparable, more sophisticated methods for Bayesian inference in LLMs. We further show that our method exhibits greater robustness against distribution shift, as reflected in its improved performance on out-of-distribution tasks. |
| title | Gaussian Stochastic Weight Averaging for Bayesian Low-Rank Adaptation of Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2405.03425 |