Post-training is the decisive step for converting a pretrained video generator into a production-oriented model that is instruction-following, controllable, and robust over long temporal horizons. This report presents a systematical post-training framework that organizes supervised policy shaping, reward-driven reinforcement learning, and preference-based refinement into a single stability-constrained optimization stack. The framework is designed around practical video-generation constraints, including high rollout cost, temporally compounding failure modes, and feedback that is heterogeneous, uncertain, and often weakly discriminative. By treating optimization as a staged, diagnostic-driven process rather than a collection of isolated tricks, the report summarizes a cohesive recipe for improving perceptual fidelity, temporal coherence, and prompt adherence while preserving the controllability established at initialization. The resulting framework provides a clear blueprint for building scalable post-training pipelines that remain stable, extensible, and effective in real-world deployment settings.
@article{arxiv.2602.07595,
title = {TeleBoost: A Systematic Alignment Framework for High-Fidelity, Controllable, and Robust Video Generation},
author = {Yuanzhi Liang and Xuan'er Wu and Yirui Liu and Yijie Fang and Yizhen Fan and Ke Hao and Rui Li and Ruiying Liu and Ziqi Ni and Peng Yu and Yanbo Wang and Haibin Huang and Qizhen Weng and Chi Zhang and Xuelong Li},
journal= {arXiv preprint arXiv:2602.07595},
year = {2026}
}