Skip to main content
Home/Risks/Gipiškis2024/Benchmarking (Cross-lingual data contamination)

Benchmarking (Cross-lingual data contamination)

Sub-category
Risk Domain

Inadequate regulatory frameworks and oversight mechanisms that fail to keep pace with AI development, leading to ineffective governance and the inability to manage AI risks appropriately.

"Models that have been trained on data encoded in multiple languages, such as LLMs trained on web-crawled data, may contain contamination that is obscured by translation [226]. The most basic form of this is when a benchmark is trans- lated to another language and then fed to the model as training data. The fact that the benchmark is translated before becoming training data can obscure the contamination from detection methods, giving false assurance that the model has generalized on the capabilities that the benchmark tests for."(p. 19)

Other risks from Gipiškis2024 (144)