SOURCE-LINKED INTELLIGENCE

Preference Data Selection for Mitigating the Alignment Tax in Large Language Models

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Aligning large language models to human preferences is crucial for real-world deployment but frequently incurs an alignment tax, leading to the catastrophic forgetting of pre-trained general capabilities. While previous works primarily frame this problem as an optimization or architectural challenge, the inherent characteristics of preference data that drive this degradation remain largely underexplored. In this paper, we propose BALIGN, a balanced data selection strategy that explicitly mitigates catastrophic forgetting while optimizing alignment efficacy. Through theoretical and empirical an

Read original source ↗ Open in workspace

recordType

paper

region

Global