@misc{indiciaedb768e2836ee,
  title = {Forge-and-Quench: Enhancing Image Generation for Higher Fidelity in Unified Multimodal Models},
  author = {Yanbing Zeng and Jia Wang and Hanghang Ma and Junqiang Wu and Jie Zhu and Xiaoming Wei and Jie Hu},
  year = {2026},
  url = {https://arxiv.org/abs/2601.04706},
  note = {Source identifier: 2601.04706}
}