Safety alignment in large language models (LLMs) is commonly implemented as a single static policy embedded in model parameters. However, real-world deployments often require context-dependent safety rules that vary across users, regions, and applications. Existing approaches struggle to provide such conditional control: parameter-level alignment entangles safety behaviors with general capabilities, while prompt-based methods rely on natural language instructions that provide weak enforcement. We propose MOSAIC, a modular framework that enables compositional safety alignment through learnable control tokens optimized over a frozen backbone model. Each token represents a safety constraint and can be flexibly activated and composed at inference time. To train compositional tokens efficiently, we introduce order-based task sampling and a distribution-level alignment objective that mitigates over-refusal. Experiments show that MOSAIC achieves strong defense performance with substantially lower over-refusal while preserving model utility.
@article{arxiv.2603.16210,
title = {MOSAIC: Composable Safety Alignment with Modular Control Tokens},
author = {Jingyu Peng and Hongyu Chen and Jiancheng Dong and Maolin Wang and Wenxi Li and Yuchen Li and Kai Zhang and Xiangyu Zhao},
journal= {arXiv preprint arXiv:2603.16210},
year = {2026}
}