Natalie Mackraz

Evaluating the Evaluators: Investigating LLM Judges for Personalized Writing Style Assessment

Evaluating Gender Bias Transfer between Pre-trained and Prompt-Adapted Language Models

PREDICT: Preference Reasoning by Evaluating Decomposed preferences Inferred from Candidate Trajectories

Aligning LLMs by Predicting Preferences from User Writing Samples

Is Your Model Fairly Certain? Uncertainty-Aware Fairness Evaluation for LLMs