Aligning generative real-world image super-resolution models with human visual preference is challenging due to the perception--fidelity trade-off and diverse, unknown degradations. Prior approaches rely on offline preference optimization and static metric aggregation, which are often non-interpretable and prone to pseudo-diversity under strong conditioning. We propose OARS, a process-aware online alignment framework built on COMPASS, a MLLM-based reward that evaluates the LR to SR transition by jointly modeling fidelity preservation and perceptual gain with an input-quality-adaptive trade-off. To train COMPASS, we curate COMPASS-20K spanning synthetic and real degradations, and introduce a three-stage perceptual annotation pipeline that yields calibrated, fine-grained training labels. Guided by COMPASS, OARS performs progressive online alignment from cold-start flow matching to full-reference and finally reference-free RL via shallow LoRA optimization for on-policy exploration. Extensive experiments and user studies demonstrate consistent perceptual improvements while maintaining fidelity, achieving state-of-the-art performance on Real-ISR benchmarks.
@article{arxiv.2603.12811,
title = {OARS: Process-Aware Online Alignment for Generative Real-World Image Super-Resolution},
author = {Shijie Zhao and Xuanyu Zhang and Bin Chen and Weiqi Li and Qunliang Xing and Kexin Zhang and Yan Wang and Junlin Li and Li Zhang and Jian Zhang and Tianfan Xue},
journal= {arXiv preprint arXiv:2603.12811},
year = {2026}
}