研究：微调数据选择重覆盖而非难度
Rethinking Data Selection: The Importance of Coverage over Difficulty in LLM Fine-Tuning
微调前选数据必看：论文证明按难度选样本常不如随机抽，聚类选覆盖更广的数据反而更强，直接能抄的做法。
论文重新审视了 LLM 微调中的数据选择方法，发现基于难度的打分在微调中往往跑不赢随机选择。分析显示随机选择胜在覆盖度更接近真实输入分布。作者据此实现了一个基于聚类的选择方法，在多个单域任务上达到或超过随机选择，在多域的 Tülu 3 数据混合上超越所有对比方法。
Rethinking Data Selection: The Importance of Coverage over Difficulty in LLM Fine-Tuning
Selecting high-quality training data can reduce computation cost for LLM fine-tuning. Prior work on data selection developed a variety of difficulty-based scores, motivated by the intuition that harder examples are more informative. While effective for smaller classification models, we find that such methods often fail to outperform random selection for LLM fine-tuning. Our analysis reveals that random selection performs better because it is more representative of the underlying input space---i.e., has better coverage. To validate the importance of coverage, we implement a simple clustering-based selection method which selects data that is more representative of the underlying input distribution. On a set of diverse single-domain tasks, this method yields performance at or above that of random selection. On the diverse, multi-domain Tülu 3 mixture, where random sampling no longer guarantees a balanced subset, our method outperforms all methods. Overall, our results indicate that prioritizing coverage over difficulty leads to more effective data selection for LLM fine-tuning.
