Evaluating and Mitigating Linguistic Discrimination in Large Language Models

arXiv:2404.18534v1 Announce Type: new
Abstract: By training on text in various languages, large language models (LLMs) typically possess multilingual support and demonstrate remarkable capabilities in solving tasks described in different languages. However, LLMs can exhibit linguistic discrimination due to the uneven distribution of training data across languages. That is, LLMs are hard to keep the consistency of responses when faced with the same task but depicted in different languages.
In this study, we first explore the consistency in the LLMs' …

abstract arxiv capabilities cs.ai cs.cl cs.cr cs.se data discrimination distribution however language language models languages large language large language models llms multilingual support tasks text training training data type