Chenyang Si (@scy994) / X

Chenyang Si

99 posts

Chenyang Si

@scy994

Associate Professor @NJU1902 | Research Fellow @MMLabNTU | Research Scientist @SeaAIL

China

Joined May 2022

Pinned
Chenyang Si
@scy994
Nov 30, 2023
🚀🚀Now combining #FreeU with #SDXL, #ControlNet, #LCM, #ScaleCrafter, #Dreambooth, and #Animatediff, you can enhance the generation quality for free! -Project Page: chenyangsi.top/FreeU/ -Code: github.com/ChenyangSi/Fre… -Video: youtu.be/qRIx7ZVIDVQ?si…
00:00
00:02
AK
@_akhaliq
Sep 21, 2023
FreeU: Free Lunch in Diffusion U-Net paper page: huggingface.co/papers/2309.11… we uncover the untapped potential of diffusion U-Net, which serves as a "free lunch" that substantially improves the generation quality on the fly. We initially investigate the key contributions of the
Chenyang Si
@scy994
12h
🚀🚀🚀 We present AURORA-LM, a new continuous diffusion language model. We show that language can be modeled with high-capacity continuous representations and generated through diffusion. Paper updated on @HuggingPapers 📄 huggingface.co/papers/2608.02… 🌐 github.com/fyv587/AURORA-…
Chenyang Si
@scy994
13h
🚀🚀🚀 We present AURORA-LM, a new continuous diffusion language model. We show that language can be modeled with high-capacity continuous representations and generated through diffusion. 📄 arxiv.org/abs/2608.02602 🌐 github.com/fyv587/AURORA-…
Jingkang (Jake) Yang
@JingkangY
13h
🌌 AURORA-LM: a continuous-latent diffusion LM. It learns native, decodable continuous text representations and generates with diffusion. Language in continuous space — toward a unified interface with vision and action. 📄 arxiv.org/abs/2608.02602 🌐 aurora-lm-project.github.io
Chenyang Si
@scy994
Dec 21, 2025
Thanks @_akhaliq for sharing! 🎬 #LongVie2 enables continuous video generation for up to 5 minutes with: 🕹️Strong Controllability 📷 Long-term Visual Fidelity 🔒 Temporal Consistency -Paper: huggingface.co/papers/2512.13… -Page: vchitect.github.io/LongVie2-proje… -Code: github.com/Vchitect/LongV…
00:00
02:17
AK
@_akhaliq
Dec 21, 2025
LongVie 2 Multimodal Controllable Ultra-Long Video World Model
Chenyang Si
@scy994
Dec 6, 2025
📢 Our #NextVid workshop @ #NeurlPS2025 schedule is live! 🎉 @Kaleidudu @liuziwei7 👏🏻Welcome to the full program! ☀️ Many thanks to our wonderful speakers: @liu_mingyu, @dimadamen, @jiajunwu_cs, @haozhangml, @Enjoy_Yi, @xxunhuang, @sainingxie, @HengshuangZhao