Tome Eftimov

This paper introduces robustness indicators to systematically analyze how multilingual text embedding model rankings change based on dataset composition and aggregation methods, revealing that only a small subset of models maintains consistent strong performance across diverse evaluation designs.

The paper proposes using Maximum Independent Set (MIS) algorithms on similarity graphs to select a maximally diverse and non-redundant subset of prompts for LLM benchmarking, achieving consistent rankings with significant prompt reduction.