Nous Research fine-tunes Qwen3.8-27B and releases Arete-27B
Understand this story
Nous Research performed supervised fine-tuning on Alibaba's Qwen3.8-27B using about 53 million tokens and released the open-weight model Arete-27B, with the goal of removing propagandistic content from the model's answers. According to The Decoder, the share of propagandistic answers from the model dropped from 22% to 0.9%, and factual accuracy rose from 2.32 to 2.98 (out of 5). Arete-27B's weights, test dataset, and training data are publicly available on Hugging Face under the Apache 2.0 license.
AI overview based on source reports · Updated 4 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- The DecoderState propaganda in some Chinese AI models is only "a thin layer" that fine-tuning can strip away
Nous Research fine-tuned Alibaba's Qwen3.8-27B on about 53 million tokens to create Arete-27B, which cuts its propaganda rate from 22 percent to 0.9 percent while keeping general capabilities roughly intact. The team built an 803-prompt test across seven categories and judged responses with Grok 4.7, finding Western models below one percent and GLM-5.3 and DeepSeek V4.1 Flash at 15 and 16 percent. Nous concludes the bias layer in Chinese-aligned models is shallow, applied late and thin, and removable without disturbing anything underneath.
Activity for this story
- Current comparable activity
- 9
- Peak comparable activity
- 10Nov 7 21:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.