Skip to content
AI IntelligenceNov 7, 2026AI Intelligence
Article

Nous Research retrained Alibaba's Qwen3.8-27B into Arete-27B using 53 million tokens of supervised fine-tuning…

…dropping its pro-Chinese propaganda rate from 22% to under 1% while general capabilities remained intact in tests. The team writes the political bias is "applied late, it is applied thin."

AI-generated: summaries written by AI from the linked sources. How we use AI

AI-generatedSource: The Decoder
01

Source Brief

Nous Research retrained Alibaba's Qwen3.8-27B into Arete-27B using 53 million tokens of supervised fine-tuning, dropping its pro-Chinese propaganda rate from 22% to under 1% while general capabilities remained intact in tests. The team writes the political bias is "applied late, it is applied thin."