美国 AI 实验室 Nous Research 发布 Arete-27B,直接拿阿里的 Qwen3.8-27B 当底座,专门训练它认同「西方价值观」。
Nous 承认,中国开源模型目前实力最强,但认为它们在政治敏感问题上过于贴近中国官方立场。于是,团队用约 5300 万 Token 重新微调 Qwen,重点涉及言论自由、民主、隐私等七类议题。
按照 Nous 自己设计的 803 道测试,模型回答中的「宣传」比例从 22% 降至 0.9%,通用能力基本没有下降。
有意思的是,这个「宣传率」只统计模型是否认同中国官方叙事,并不检查它是否照搬美国或其他西方国家的政治立场。测试题和评分标准由 Nous 制定,最后请马斯克旗下 xAI 的 Grok 4.7 当裁判。
目前模型权重、训练数据和评测题目均已公开。
Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values.
Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line.
After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs.
The values in an open model are one fine-tune away.
Our eval, weights, and training data are available now along with the full blog post:
顯示更多