LLMs are obsessed with Japan: 24-language study traces cultural bias to SFT
alex_verem · x · 2026-10-26
Researchers from the University of the Basque Country and Cardiff University built the CROQ dataset (culture-related open questions in 24 languages, 11 topics) and found LLMs from GPT-4o mini to Claude 3.5 Haiku show a clear tendency to answer with Japan. Notably, the bias first emerges after supervised fine-tuning, not during pre-training.
More from Research
- Open textbook on robot learning launches with chapters on control and imitation to VLA — GlenBerseth · 2026-10-26
- A $5M plan to test whether 40Hz music could treat or prevent Alzheimer's — jajoosam · 2026-10-26
- Judea Pearl Notes His Book 'Causality' Has 'Infinite Lifespan' With 5,000+ Citations — yudapearl · 2026-10-26
- Animating brain tractography with a traveling light pulse: the simple frame-length formula — Dr_Alex_Crimi · 2026-10-26
- Clinicians over-rely on wrong AI findings and under-rely on correct ones, Lancet study — zakkohane · 2026-10-26
- Depth-Grown and Looped LMs Learn In-Context Learning Earlier Than Dense Models — omarsar0 · 2026-10-26