Exploring Explanations Improves the Robustness of In-Context Learning
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_ | 1866912410943619072 |
|---|---|
| author | Honda, Ukyo Oka, Tatsushi |
| author_facet | Honda, Ukyo Oka, Tatsushi |
| contents | In-context learning (ICL) has emerged as a successful paradigm for leveraging large language models (LLMs). However, it often struggles to generalize beyond the distribution of the provided demonstrations. A recent advancement in enhancing robustness is ICL with explanations (X-ICL), which improves prediction reliability by guiding LLMs to understand and articulate the reasoning behind correct labels. Building on this approach, we introduce an advanced framework that extends X-ICL by systematically exploring explanations for all possible labels (X$^2$-ICL), thereby enabling more comprehensive and robust decision-making. Experimental results on multiple natural language understanding datasets validate the effectiveness of X$^2$-ICL, demonstrating significantly improved robustness to out-of-distribution data compared to the existing ICL approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_02378 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Exploring Explanations Improves the Robustness of In-Context Learning Honda, Ukyo Oka, Tatsushi Computation and Language Artificial Intelligence Machine Learning In-context learning (ICL) has emerged as a successful paradigm for leveraging large language models (LLMs). However, it often struggles to generalize beyond the distribution of the provided demonstrations. A recent advancement in enhancing robustness is ICL with explanations (X-ICL), which improves prediction reliability by guiding LLMs to understand and articulate the reasoning behind correct labels. Building on this approach, we introduce an advanced framework that extends X-ICL by systematically exploring explanations for all possible labels (X$^2$-ICL), thereby enabling more comprehensive and robust decision-making. Experimental results on multiple natural language understanding datasets validate the effectiveness of X$^2$-ICL, demonstrating significantly improved robustness to out-of-distribution data compared to the existing ICL approaches. |
| title | Exploring Explanations Improves the Robustness of In-Context Learning |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.02378 |