R-Select: A Robust Multi-Metric Data Selection Approach for Fine-Tuning Large Language Models
Xin Gao, Xiaoyang Wang, Yun Zhu, Zheng Liu, Conghui He, Lijun Wu
Abstract
The transition from architecture-centric scaling to data-centric refinement has established high-quality data as a critical determinant of Large Language Model performance, particularly for complex reasoning and instruction following. However, effective data selection remains a persistent bottleneck: simple heuristic filters often fail to capture multifaceted data features (e.g., reasoning depth and diversity), while advanced model-based scoring methods typically prioritize isolated quality dimensions, failing to provide a holistic assessment. To address these challenges, we propose R-Select, a robust and scalable framework that optimizes data selection with 30 distinct quality metrics. Recognizing that optimizing such a high-dimensional feature space is non-trivial, R-Select introduces a novel hierarchical optimization strategy. This approach structurally decomposes the search problem by first clustering correlated metrics into functional groups based on statistical dependencies. It then executes a two-stage optimization process: performing intra-group refinement to maximize local representational power, followed by inter-group integration to balance global quality domains. Crucially, to ensure computational efficiency, R-Select employs a low-resource proxy strategy, utilizing a lightweight model on a small data subset to learn an optimal selection policy that is transferable to the target model. Extensive experiments demonstrate that R-Select consistently outperforms both heuristic baselines and model-based methods, offering a robust solution for high-quality data curation.
Ask Lune
Know where this paper stands in minutes, not an afternoon.
Lune can read the top-tier papers that cite it and the ones closest to it, and tells you what it claims, what came after and where it falls short, each point cited to the paper it comes from. No skimming, no AI slop.
Related papers
- Meta-rater: A Multi-dimensional Data Selection Method for Pre-training Language ModelsXinlin Zhuang, Jiahui Peng, Ren Ma, Yinfan Wang et al.ACL 2025 · 15 citations
- Select2Reason: Efficient Instruction-Tuning Data Selection for Long-CoT ReasoningCehao Yang, Xueyuan Lin, Xiaojun Wu, Chengjin Xu et al.ACL Findings 2026 · 1 citation
- Towards High Data Efficiency in Reinforcement Learning with Verifiable RewardXinyu Tang, Zhenduo Zhang, Yurou Liu, Xin Zhao et al.ICLR 2026 · 18 citations
- FIRE: Flexible Integration of Data Quality Ratings for Effective PretrainingLiangyu Xu, Xuemiao Zhang, Feiyu Duan, Sirui Wang et al.EMNLP 2025
- Scale Down to Speed Up: Dynamic Data Selection for Reinforcement LearningZhuoyue Chen, Jihai Zhang, Ben Liu, Fangquan Lin et al.EMNLP Findings 2025 · 1 citation
