New paper with Dr. Ahmet Kurnaz. Geopolitical bias attributed to LLMs may be a result of the linguistic properties of training data rather than political biases of countries that operate them ieeexplore.ieee.org/abstract/doc...
ieeexplore.ieee.org
Can We Trust LLMs? Inter-model Reliability and Geopolitical Bias under Zero-Shot Classification
This study analyzes the classification consistency of Large Language Models (LLMs) using geopolitical texts collected from social media. Data gathered under the headings of the United States, Russia, ...