Uncovering inequalities in new knowledge learning by large language models across different languages
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909527241129984 |
|---|---|
| author | Wang, Chenglong Tang, Haoyu Yang, Xiyuan Xie, Yueqi Suh, Jina Sitaram, Sunayana Huang, Junming Xie, Yu Gong, Zhaoya Xie, Xing Wu, Fangzhao |
| author_facet | Wang, Chenglong Tang, Haoyu Yang, Xiyuan Xie, Yueqi Suh, Jina Sitaram, Sunayana Huang, Junming Xie, Yu Gong, Zhaoya Xie, Xing Wu, Fangzhao |
| contents | As large language models (LLMs) gradually become integral tools for problem solving in daily life worldwide, understanding linguistic inequality is becoming increasingly important. Existing research has primarily focused on static analyses that assess the disparities in the existing knowledge and capabilities of LLMs across languages. However, LLMs are continuously evolving, acquiring new knowledge to generate up-to-date, domain-specific responses. Investigating linguistic inequalities within this dynamic process is, therefore, also essential. In this paper, we explore inequalities in new knowledge learning by LLMs across different languages and four key dimensions: effectiveness, transferability, prioritization, and robustness. Through extensive experiments under two settings (in-context learning and fine-tuning) using both proprietary and open-source models, we demonstrate that low-resource languages consistently face disadvantages across all four dimensions. By shedding light on these disparities, we aim to raise awareness of linguistic inequalities in LLMs' new knowledge learning, fostering the development of more inclusive and equitable future LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_04064 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Uncovering inequalities in new knowledge learning by large language models across different languages Wang, Chenglong Tang, Haoyu Yang, Xiyuan Xie, Yueqi Suh, Jina Sitaram, Sunayana Huang, Junming Xie, Yu Gong, Zhaoya Xie, Xing Wu, Fangzhao Computation and Language Artificial Intelligence Computers and Society As large language models (LLMs) gradually become integral tools for problem solving in daily life worldwide, understanding linguistic inequality is becoming increasingly important. Existing research has primarily focused on static analyses that assess the disparities in the existing knowledge and capabilities of LLMs across languages. However, LLMs are continuously evolving, acquiring new knowledge to generate up-to-date, domain-specific responses. Investigating linguistic inequalities within this dynamic process is, therefore, also essential. In this paper, we explore inequalities in new knowledge learning by LLMs across different languages and four key dimensions: effectiveness, transferability, prioritization, and robustness. Through extensive experiments under two settings (in-context learning and fine-tuning) using both proprietary and open-source models, we demonstrate that low-resource languages consistently face disadvantages across all four dimensions. By shedding light on these disparities, we aim to raise awareness of linguistic inequalities in LLMs' new knowledge learning, fostering the development of more inclusive and equitable future LLMs. |
| title | Uncovering inequalities in new knowledge learning by large language models across different languages |
| topic | Computation and Language Artificial Intelligence Computers and Society |
| url | https://arxiv.org/abs/2503.04064 |