Label-Efficient Model Selection for Text Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910473827385344 |
|---|---|
| author | Ashury-Tahan, Shir Gera, Ariel Sznajder, Benjamin Choshen, Leshem Ein-Dor, Liat Shnarch, Eyal |
| author_facet | Ashury-Tahan, Shir Gera, Ariel Sznajder, Benjamin Choshen, Leshem Ein-Dor, Liat Shnarch, Eyal |
| contents | Model selection for a given target task can be costly, as it may entail extensive annotation of the quality of outputs of different models. We introduce DiffUse, an efficient method to make an informed decision between candidate text generation models based on preference annotations. DiffUse reduces the required amount of annotations, thus saving valuable time and resources in performing evaluation. DiffUse intelligently selects instances by clustering embeddings that represent the semantic differences between model outputs. Thus, it is able to identify a subset of examples that are more informative for preference decisions. Our method is model-agnostic, and can be applied to any text generation model for selecting between models, prompts and configurations. Moreover, we propose a practical iterative approach for dynamically determining how many instances to annotate. In a series of experiments over hundreds of model pairs, we demonstrate that DiffUse can dramatically reduce the required number of annotations -- by up to 75% -- while maintaining high evaluation reliability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_07891 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Label-Efficient Model Selection for Text Generation Ashury-Tahan, Shir Gera, Ariel Sznajder, Benjamin Choshen, Leshem Ein-Dor, Liat Shnarch, Eyal Computation and Language Machine Learning Model selection for a given target task can be costly, as it may entail extensive annotation of the quality of outputs of different models. We introduce DiffUse, an efficient method to make an informed decision between candidate text generation models based on preference annotations. DiffUse reduces the required amount of annotations, thus saving valuable time and resources in performing evaluation. DiffUse intelligently selects instances by clustering embeddings that represent the semantic differences between model outputs. Thus, it is able to identify a subset of examples that are more informative for preference decisions. Our method is model-agnostic, and can be applied to any text generation model for selecting between models, prompts and configurations. Moreover, we propose a practical iterative approach for dynamically determining how many instances to annotate. In a series of experiments over hundreds of model pairs, we demonstrate that DiffUse can dramatically reduce the required number of annotations -- by up to 75% -- while maintaining high evaluation reliability. |
| title | Label-Efficient Model Selection for Text Generation |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2402.07891 |