Rethinking Data Selection at Scale: Random Selection is Almost All You Need
This work addresses data selection inefficiencies for practitioners fine-tuning LLMs, though it is incremental as it builds on existing selection methods by testing them at larger scales.
The paper tackled the problem of selecting training data for supervised fine-tuning of large language models at scale, finding that random selection often matches or outperforms complex methods on million-scale datasets, and that diversity and token-length filtering improve results, with specific gains like enhanced performance for weaker models like Llama3.
Supervised fine-tuning (SFT) is crucial for aligning Large Language Models (LLMs) with human instructions. The primary goal during SFT is to select a small yet representative subset of training data from the larger pool, such that fine-tuning with this subset achieves results comparable to or even exceeding those obtained using the entire dataset. However, most existing data selection techniques are designed for small-scale data pools, which fail to meet the demands of real-world SFT scenarios. In this paper, we replicated several self-scoring methods those that do not rely on external model assistance on two million scale datasets, and found that nearly all methods struggled to significantly outperform random selection when dealing with such large-scale data pools. Moreover, our comparisons suggest that, during SFT, diversity in data selection is more critical than simply focusing on high quality data. We also analyzed the limitations of several current approaches, explaining why they perform poorly on large-scale datasets and why they are unsuitable for such contexts. Finally, we found that filtering data by token length offers a stable and efficient method for improving results. This approach, particularly when training on long text data, proves highly beneficial for relatively weaker base models, such as Llama3.