The community keeps ignoring the Chinese ML ecosystem work. They are doing amazing stuff with interesting LLMs, VLMs, audio, and diffusion models 👀 Qwen, Yi, DeepSeek, Yuan, WizardLM, ChatGLM, CogVLM, Baichuan, InternLM, OpenBMB, Skywork, ChatTTS, Ernie, HunyuanDiT, etc.

#cvpr2024 #seattle CVPR day 1. Pretty nice organizations make me enjoy a lot in communication with passinate researchers worldwide. Thanks for all!

Kai Wang ✍️Also, Kai Wang presented his poster: "LocInv: Localization-aware Inversion for Text-Guided Image Editing".

🌈 Exciting news! Learn about ColorPeel's innovative approach to color prompt learning with diffusion models in T2I generation. Achieve precise color matching and disentangle color and shape with our method. Check out the blog post at bit.ly/4bDkQwT #T2I #ColorPeel

ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement. arxiv.org/abs/2407.07197

Congrats! Excellent work!

🚨ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement [ECCV'24] 🌟𝐏𝐫𝐨𝐣: moatifbutt.github.io/colorpeel/ 🚀𝐀𝐛𝐬: arxiv.org/abs/2407.07197 learn specific color prompts tailored to user-selected colors to tackle color prompt learning task

ColorPeel can generate objects in images with specific colors and shapes. Links ⬇️

Introducing Meta Segment Anything Model 2 (SAM 2) — the first unified model for real-time, promptable object segmentation in images & videos. SAM 2 is available today under Apache 2.0 so that anyone can use it to build their own experiences Details ➡️ go.fb.me/p749s5

Black Myth: Wukong has sold 10 million copies across all platforms. (Data as of 21:00 Beijing time, August 23, 2024) Thanks to all players worldwide for your support and love. Have a great gaming weekend! #BlackMythWukong