Determine cross-linguistic iteration requirements

Determine whether low-resource languages such as Estonian and Finnish require more repeated queries than high-resource languages such as English and German to achieve equivalent measurement precision in cross-linguistic large-language-model audits.

Background

The cross-linguistic analysis finds that model-related stability variance is substantially larger than language-related variance, while also identifying a non-negligible model-by-language interaction driven mainly by lower stability for Gemini on Estonian and Finnish.

The authors state that the interaction between language and iteration count has not been characterized. Resolving this issue is important for equitable audit design because a common iteration count could provide different levels of measurement precision across languages.

References

Specifically, whether low-resource languages (Estonian, Finnish) require more iterations than high-resource languages (English, German) to achieve equivalent measurement precision is an open empirical question with implications for equitable cross-linguistic auditing.

— The Dice Roll Method: A Standardized Protocol for Repeated-Query Auditing of Large Language Model Brand Recommendations  (2609.04047 - Żatuchin, 3 Sep 2026) in Section 11, subsection “Future Work”