Published June 3, 2026 | Version v1.0.4-thesis
Software Open

Assessing the Reliability of LLM Annotations in the Context of Demographic Bias and Model Explanation

Description

Code and analysis comparing human annotators of varied demographics against demographically persona-prompted LLMs, using SHAP to attribute label decisions to specific text spans. Archived snapshot for the v1.0-thesis release accompanying the doctoral thesis Let Me Explain! Explainable NLP for Understanding Large Language Models (Utrecht University, 2026).

Files

Files (223.0 kB)

Name Size Download all
md5:dde29fcb5c5936b373644e96faf665bd
223.0 kB Download

Additional details

Related works

Is supplement to
10.18653/v1/2025.gebnlp-1.9 (DOI)