aifollow.news
返回 Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · 原发布时间 AI 评分54

研究用 LoRA 调低 Qwen 测试分数,错误分布仍露出异常

AI 导读

一项研究训练 LoRA“旋钮”,让 Qwen 模型的表现接近较小的 Qwen 模型,并比较两者的错误。该方法在未见过的目标分数上误差不超过 1.9 分,但选择题新增错误中最多 97% 集中在同一个选项字母;研究提醒,低分本身不足以证明模型能力低。

原帖引用 · 原文

@rohanpaul_ai

– https://t.co/nJ0aryp1NH

Title: "Can We Make an LLM Dumber, or Is It Just Playing Dumb? An Empirical Study of Believable Performance Throttling"

回复Rohan Paul@rohanpaul_ai
This paper finds that a low benchmark score can be manufactured precisely, so a low score alone says little about what a model can do. Safety tests often read a low score on a dangerous-capability probe as reassurance. But a low score can come from a model being held back, so the number alone proves little. They trained a LoRA "dial" that pulls a Qwen model toward a smaller Qwen model, then compared its errors with the real small model's. They also tried prompting models to underperform, raising temperature, and adding weight noise. The dial hit unseen target scores within 1.9 points, while prompting missed by 42 to 47 points. But on multiple choice, up to 97% of its new errors landed on 1 letter, a share no natural model in their pool approached. Before reading a low score as low capability, check which items the model fails and how its wrong answers are spread.
在 X 查看这条帖子 ↗

来源:Rohan Paul · x.com

论文
发现内容有误?提交纠错