Models for paper "Relevant for the Right Reasons? Investigating Lexical Biases in Re-ranking with Large Language Models" (EMNLP MRL workshop 2025).