Update README.md

README.md

@@ -18,12 +18,6 @@ JanusFlow introduces a minimalist architecture that integrates autoregressive
 language models with rectified flow, a state-of-the-art method in generative modeling. Our
 key finding demonstrates that rectified flow can be straightforwardly trained within the large
 language model framework, eliminating the need for complex architectural modifications.
-To further improve the performance of our unified model, we adopt two key strategies: (i)
-decoupling the understanding and generation encoders, and (ii) aligning their representations
-during unified training. Extensive experiments show that JanusFlow achieves comparable or
-superior performance to specialized models in their respective domains, while significantly
-outperforming existing unified approaches across standard benchmarks. This work represents
-a step toward more efficient and versatile vision-language models.
 
 [JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation](TBD)
 
@@ -36,12 +30,12 @@ a step toward more efficient and versatile vision-language models.
 
 ### 2. Model Summary
 
-Janus is a unified understanding and generation MLLM, which decouples visual encoding for multimodal understanding and generation. 
-Janus is constructed based on the DeepSeek-LLM-1.3b-base which is trained on an approximate corpus of 500B text tokens.
-For multimodal understanding, it uses the [SigLIP-L](https://huggingface.co/timm/ViT-L-16-SigLIP-384) as the vision encoder, which supports 384 x 384 image input. For image generation, Janus uses the tokenizer from [here](https://github.com/FoundationVision/LlamaGen) with a downsample rate of 16.
+JanusFlow is a unified understanding and generation MLLM, which decouples visual encoding for multimodal understanding and generation, which is constructed based on DeepSeek-LLM-1.3b-base.
+For multimodal understanding, it uses the [SigLIP-L](https://huggingface.co/timm/ViT-L-16-SigLIP-384) as the vision encoder, which supports 384 x 384 image input. 
+For image generation, JanusFlow uses rectified flow to generate in the latent space of [SDXL-VAE](https://huggingface.co/stabilityai/sdxl-vae).
 
 <div align="center">
-<img alt="image" src="arch.jpg" style="width:90%;">
+<img alt="image" src="arch.png" style="width:90%;">
 </div>
 
 ## 3. Quick Start
@@ -51,19 +45,12 @@ Please refer to [**Github Repository**](https://github.com/deepseek-ai/Janus)
 
 ## 4. License
 
-This code repository is licensed under [the MIT License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-CODE). The use of Janus models is subject to [DeepSeek Model License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-MODEL).
+This code repository is licensed under [the MIT License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-CODE). The use of JanusFlow models is subject to [DeepSeek Model License](https://github.com/deepseek-ai/DeepSeek-LLM/blob/HEAD/LICENSE-MODEL).
+
 ## 5. Citation
 
 ```
-@misc{wu2024janus,
-      title={Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation}, 
-      author={Chengyue Wu and Xiaokang Chen and Zhiyu Wu and Yiyang Ma and Xingchao Liu and Zizheng Pan and Wen Liu and Zhenda Xie and Xingkai Yu and Chong Ruan and Ping Luo},
-      year={2024},
-      eprint={2410.13848},
-      archivePrefix={arXiv},
-      primaryClass={cs.CV},
-      url={https://arxiv.org/abs/2410.13848}, 
-}
+TBD
 ```
 
 ## 6. Contact