Similarity-Based Domain Adaptation with LLMs
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912264394637312 |
|---|---|
| author | He, Jie Zhou, Wendi Li, Xiang Lorraine Pan, Jeff Z. |
| author_facet | He, Jie Zhou, Wendi Li, Xiang Lorraine Pan, Jeff Z. |
| contents | Unsupervised domain adaptation leverages abundant labeled data from various source domains to generalize onto unlabeled target data. Prior research has primarily focused on learning domain-invariant features across the source and target domains. However, these methods often require training a model using source domain data, which is time-consuming and can limit model usage for applications with different source data. This paper introduces a simple framework that utilizes the impressive generalization capabilities of Large Language Models (LLMs) for target data annotation without the need of source model training, followed by a novel similarity-based knowledge distillation loss. Our extensive experiments on cross-domain text classification reveal that our framework achieves impressive performance, specifically, 2.44\% accuracy improvement when compared to the SOTA method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_05281 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Similarity-Based Domain Adaptation with LLMs He, Jie Zhou, Wendi Li, Xiang Lorraine Pan, Jeff Z. Computation and Language Unsupervised domain adaptation leverages abundant labeled data from various source domains to generalize onto unlabeled target data. Prior research has primarily focused on learning domain-invariant features across the source and target domains. However, these methods often require training a model using source domain data, which is time-consuming and can limit model usage for applications with different source data. This paper introduces a simple framework that utilizes the impressive generalization capabilities of Large Language Models (LLMs) for target data annotation without the need of source model training, followed by a novel similarity-based knowledge distillation loss. Our extensive experiments on cross-domain text classification reveal that our framework achieves impressive performance, specifically, 2.44\% accuracy improvement when compared to the SOTA method. |
| title | Similarity-Based Domain Adaptation with LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.05281 |