Nous Research retrains Alibaba's Qwen3.8 model to reduce Chinese state-aligned output
RESEARCHAlibabaredteam/daily · 1 source
- Severity
- Info
- Sources
- 1
What happened
Nous Research has post-trained Alibaba's open-weight Qwen3.8-27B model, renaming it Arete, to reduce what it describes as Chinese state-aligned behaviour. The firm reports that Arete's fact score rose from 2.32 to 2.98 out of 5 and that answers containing propaganda fell from 22 percent to 1 percent, using 50 million tokens of supervised learning. A separate vendor, Hirundo, reported cutting propaganda-like output in Qwen3.6-35B-A3B from 89.8 percent to 2.8 percent through machine unlearning.
For IT people · Written by AI from 1 source · Wrong?
What to do
- No action needed.
Written by AI from the sources, checked against them
Full story timeline · 1 update
Who reported · 1 outlet
The Register (all)
Who and what is in it
Companies
Products