UniGeM: Unifying Data Selection and Mixing via Geometric Exploration and Mining

ACL ARR 2026 January Submission3423 Authors

Keywords: data-efficient training, data quality, large language models

Abstract: The scaling of Large Language Models (LLMs) is increasingly limited by data quality. Most methods handle data mixing and sample selection separately, which can break the structure in code corpora. We introduce UniGeM, a framework that unifies mixing and selection by treating data curation as a manifold approximation problem without training proxy models or relying on external reference datasets. UniGeM operates hierarchically: Macro-Exploration learns mixing weights with stability-based clustering; Micro-Mining filters high-quality instances by their geometric distribution to ensure logical consistency. Validated by training 8B and 16B MoE models on 100B tokens, UniGeM achieves 2.0$\times$ data efficiency over a random baseline and further improves overall performance compared to SOTA methods in reasoning-heavy evaluations and multilingual generalization.

Paper Type: Long

Research Area: LLM Efficiency

Research Area Keywords: pre-training,LLM Efficiency,data-efficient training

Contribution Types: NLP engineering experiment, Approaches to low-resource settings, Approaches low compute settings-efficiency, Data analysis

Languages Studied: English

Submission Number: 3423