Kai Chen
@kaichen1998
- 🤔New reason LLMs keep releasing before your MLLM alignment finish? Try RACRO! Train once, flexible change to novel LLM reasoners when inference! 🔥Paper: arxiv.org/abs/2506.04559 🔥Code: github.com/gyhdog99/RACRO2 🔥@Gradio Demo: huggingface.co/spaces/Emova-o… #multimodalai #CVPR #CVPR25
- 🔥Check our @CVPR poster @ 10:30 am on June 14, Hall D #47! EMOVA is a fully open-sourced E2E Omni-modal LLM with SoTA vision-language and speech results! 🔥 Web: emova-ollm.github.io 🔥 Code: github.com/emova-ollm/EMO… 🔥 Demo: huggingface.co/spaces/Emova-o… #CVPR2025 #CVPR25 #CVPR
- 🔥Check our @CVPR poster @ 10:30 am on June 14, Hall D #47! EMOVA is a fully open-sourced E2E Omni-modal LLM with SoTA vision-language and speech results! 🔥 Web: emova-ollm.github.io 🔥 Code: github.com/emova-ollm/EMO… 🔥 Demo: huggingface.co/spaces/Emova-o… #CVPR2025 #CVPR25 #CVPREMOVA Empowering Language Models to See, Hear and Speak with Vivid Emotions discuss: huggingface.co/papers/2409.18… GPT-4o, an omni-modal model that enables vocal conversations with diverse emotions and tones, marks a milestone for omni-modal foundation models. However, empowering
- 🎉Glad to see progress for Omni LLM @Alibaba_Qwen! 🤔Want to train your own Omni LLM? Consider our #CVPR2025 EMOVA as an open-source alternative! All datasets, code (train/infer), and weights (3B/7B/72B) are available! @_akhaliq @huggingface @CVPR @CVPRConf #Qwen #MultimodalVoice Chat + Video Chat! Just in Qwen Chat (chat.qwen.ai)! You can now chat with Qwen just like making a phone call or making a video call! Check the demo in youtube.com/watch?v=yKcANd… What's more, we opensource the model behind all this, Qwen2.5-Omni-7B, under the
