AI 用希腊语思考，你根本看不见
你让 AI 用希腊语回答问题，它其实在用英语想——你读不到、改不了、也查不了。这篇把三个前沿 MoE 模型(阿里、OpenAI、英伟达，各 3.6-4B 活跃参数)微调成用希腊语推理。准确率基准几乎没变，而且基准本身噪音大：换个随机种子就波动 7.7 分，比所有数据和配方的影响都大。真正的变化在准确率看不见的地方：基座模型 1000 次推理里 0 次用希腊语，即使用希腊语提问；监督微调后，98% 的推理用提问语言，一个模型还省了 3 倍 token，语法评分提升，通用能力几乎不变。但监督微调修不了自己的缺陷：四分之一回答跳过指定格式，答案漏进推理通道，明确指令“用英语想”只有不到一半时间遵守。用可验证奖励的强化学习(训练前预注册)把格式回退从 24% 降到 2.5%，泄漏从 3.5% 降到 0.0%,“用英语想”遵守率提升 9.1 个百分点，而希腊语推理习惯在只优化准确率的梯度下纹丝不动。它不是你明天能用上的——但如果你在部署非英语 AI 助手、做多语言客服、或审计模型输出，这篇告诉你：准确率是面镜子，只照你想看的。
📄 原文摘要(英文)
Take three frontier mixture-of-experts models (Alibaba, OpenAI, NVIDIA; 3.6-4.0B active parameters each) and fine-tune them to reason in a low-resource language. On accuracy benchmarks almost nothing happens, and the benchmark itself is noise at this scale: changing only the random seed moves the score by 7.7 points, more than every data and recipe effect we measured. That null is our first result. The real changes live where accuracy cannot see. Base models never think in Greek: 0 of 1,000 reasoning traces, even when the question is Greek, so the model answers correctly while reasoning in a form its user cannot read, audit, or correct. After supervised fine-tuning (SFT), every released checkpoint reasons in the language of the question on ~98% of items, one family at 3x fewer tokens, with judged grammaticality improving on all four models and general ability within a few points of each base: nothing was forgotten, and fluency was gained. We propose six behavioural dimensions that make such changes measurable, each gated to reject any metric that correlates with output length, and we report how our own instruments lied: six failures, each caught by a control. What SFT cannot do is fix its own defects: a quarter of answers skip the requested format, answers leak into the reasoning channel, and an explicit "think in English" is obeyed under half the time. Reinforcement learning with verifiable rewards, pre-registered before training, fixes the first two outright (fallback 24% to 2.5%, leak 3.5% to 0.0%, both against a flat random-reward control) and moves the third (+9.1pp), while the Greek reasoning habit survives an accuracy-only gradient untouched. We release five checkpoints. The instruments, the controls and the pre-registration travel to any low-resource language; Greek is the case that let us measure them.