2 papers
cs.LG2026
It's the humans, not the data: Geopolitical bias in LLMs originates in post-training, amplified by the language of the prompt
Stuart Bladon, Brinnae Bent
It has generally been assumed that geopolitical bias in language models originates from the training data used during the pre-training phase. We tested seven open-weight LLM pairs…
q-bio.NC2026
Feature Visualization Recovers Known Cortical Selectivity from TRIBE v2
Stuart Bladon, Brinnae Bent
Brain encoder models predict cortical fMRI responses from the internal activations of pretrained vision and language networks, and are typically evaluated by held-out prediction ac…