| --- |
| license: apache-2.0 |
| language: |
| - en |
| datasets: |
| - ai9stars/Cheers-Training-Data |
| --- |
| |
| <div align="center"> |
|
|
| # ***Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation*** |
|
|
| Yichen Zhang<sup>1*</sup>, [Da Peng](https://pengda02.github.io/)<sup>2*</sup>, [Zonghao Guo](https://scholar.google.com/citations?user=h1I6LJcAAAAJ&hl=zh-CN)<sup>1†</sup>, Zijian Zhang<sup>3</sup>, Xuesong Yang<sup>3</sup>, |
|
|
| Tong Sun<sup>3</sup>, Shichu Sun<sup>3</sup>, Yidan Zhang<sup>3</sup>, Yanghao Li<sup>1</sup>, Haiyan Zhao<sup>1</sup>, Wang Xu<sup>1</sup>, |
|
|
| Qi Shi<sup>1</sup>, Yangang Sun<sup>1</sup>, Chi Chen<sup>1</sup>, Shuo Wang<sup>1</sup>, Yukun Yan<sup>1</sup>, Xu Han<sup>1</sup>, |
|
|
| Qiang Ma<sup>1</sup>, [Wei Ke](https://scholar.google.com/citations?hl=en&user=BENt-uEAAAAJ)<sup>2</sup>, Liang Wang<sup>3</sup>, Zhiyuan Liu<sup>1</sup>, Maosong Sun<sup>1</sup> |
|
|
| <sup>1</sup>Tsinghua University, |
| <sup>2</sup>Xi'an Jiaotong University, |
| <sup>3</sup>University of Chinese Academy of Sciences |
|
|
| \* Equal contribution |
| † Corresponding author |
|
|
|
|
|
|
| ## 🌟 What is ***Cheers***? |
| A recent cutting-edge topic in multimodal modeling is to unify visual comprehension and generation within a single model. However, the two tasks demand mismatched decoding regimes and visual representations, making it non-trivial to jointly optimize within a shared feature space. In this work, we present ***Cheers***, a unified multimodal model that decouples patch-level details from semantic representations, thereby stabilizing semantics for multimodal understanding and improving fidelity for image generation via gated detail residuals. ***Cheers*** includes three key components: (i) a unified vision tokenizer that encodes and compresses image latent states into semantic tokens for efficient LLM conditioning, (ii) an LLM-based Transformer that unifies autoregressive decoding for text generation and diffusion decoding for image generation, and (iii) a cascaded flow matching head that decodes visual semantics first and then injects semantically gated detail residuals from the vision tokenizer to refine high-frequency content. Experiments on popular benchmarks demonstrate that ***Cheers*** matches or surpasses advanced UMMs in both visual understanding and generation. Notably, ***Cheers*** outperforms the Tar-1.5B on the popular benchmarks GenEval and MMBench, while requiring only 20% of the training cost, indicating effective and efficient (i.e., 4x token compression) unified multimodal modeling. |
|
|
|
|
| <!-- Provide the basic links for the model. --> |
|
|
| - **Repository:** [[https://github.com/AI9Stars/Cheers](https://github.com/AI9Stars/Cheers)] |
| - **Paper:** [[https://arxiv.org/abs/2603.12793](https://arxiv.org/abs/2603.12793)] |
|
|
| </div> |
|
|
| ## Model Evaluation |
|
|
| ### compare with cheers |
|
|
|  |
|
|
|
|
| ## 🔥 News |
|
|
| - **[2026/03/19]** 🎉 [**Demo**](https://huggingface.co/spaces/prithivMLmods/Cheers-HF-Demo) is now available on Hugging Face. Thanks to [Prithiv Sakthi](https://github.com/prithivsakthiur) for setting it up! |
| - **[2026/03/16]** 📢 The ***Cheers*** paper is officially released. |
| - **[2026/03/16]** 🛠 We open-source the evaluation code and training pipeline. **Our codebase is highly efficient: training on 3.8M samples takes only about two days on a single machine with 8×A100 GPUs.** |
| - **[2026/03/16]** 📦 The model checkpoints of ***Cheers*** are now available. |
|
|
|
|
| ## Model Card Contact |
| For any questions or collaborations, feel free to contact us : ) |
|
|
| <p align="left"> |
| 📧 <a href="MetaPDa@gmail.com">MetaPDa@gmail.com</a>   |    📧 <a href="guozonghao96@outlook.com">guozonghao96@outlook.com</a>   |    📧 <a href="yichen0zhang@gmail.com">yichen0zhang@gmail.com</a>   |
| </p> |
|
|
| ## 📖 Citation |
| If you find ***Cheers*** useful, please cite ***Cheers*** technical report using this BibTeX. |
|
|
| ```bibtex |
| @article{zhang2026cheers, |
| title={CHEERS: DECOUPLING PATCH DETAILS FROM SEMANTIC REPRESENTATIONS ENABLES UNIFIED MULTIMODAL COMPREHENSION AND GENERATION}, |
| author={Zhang, Yichen and Peng, Da and Guo, Zonghao and Zhang, Zijian and Yang, Xuesong and Sun, Tong and Sun, Shichu and Zhang, Yidan and Li, Yanghao and Zhao, Haiyan and others}, |
| journal={arXiv preprint arXiv:2603.12793}, |
| year={2026} |
| } |
| ``` |
