Eunbyung Park

@silverbottlep

Assistant Professor, Yonsei University, South Korea | Ph.D., UNC Chapel Hill | Prev at @nuro @microsoft @GoogleDeepMind @MSFTResearch @AdobeResearch @hplabs

silverbottlep.github.io

Joined July 2015

Pinned
Eunbyung Park
@silverbottlep
Dec 10, 2025
🚀 Super excited to share our latest work, a scalable feed-forward 3DGS model. We've been working on feed-forward 3DGS for more than a year, this is the result of our blood, sweat, tears… and 💸(GPUs;;;) gynjn.github.io/MVP/
00:00
Eunbyung Park
@silverbottlep
Jul 21
Buckle up. South Korea is just getting started. 🇰🇷🔥
AK
@_akhaliq
Jul 21
Motif-3-Beta just dropped on Hugging Face ~314B total parameters / ~13B active per token (sparse MoE) 256K context length (262,144 tokens), natively long-context Sparse routing: 384 experts with 8 activated per token, plus 1 shared expert Multilingual, general-purpose
Eunbyung Park
@silverbottlep
Jul 3
Accepted to #ECCV2026! 🎉 We've also released the code, it should work like a charm. If it doesn't, feel free to poke @roodiiiiiiiii 😄 github.com/Ubin108/Group3D
AK
@_akhaliq
Mar 24
Group3D MLLM-Driven Semantic Grouping for Open-Vocabulary 3D Object Detection paper: huggingface.co/papers/2603.21…
00:00
Eunbyung Park
@silverbottlep
Jul 1
Code dropped! github.com/HwasikJeong/2X…
Wildminder
@wildmindai
Mar 25
Photos to 3D model with 2Xplat. Pose-free 3D Gaussian Splatting - 0.15s–0.31s inference; - beats YoNoSplat; - high-fidelity splats hwasikjeong.github.io/2Xplat/
00:00
00:00
Eunbyung Park
@silverbottlep
Mar 27
Thanks for featuring our work! 🙏 For a bit more context: open-vocab, pose-free multi-view input, no point clouds or depth, and zero-shot. Kudos to @roodiiiiiiiii👏 🔥 +17 mAP vs. prev SOTA 🔥 🏠Project: ubin108.github.io/Group3D/ 📄Paper: arxiv.org/abs/2603.21944 💻Code (coming
AK
@_akhaliq
Mar 24
Group3D MLLM-Driven Semantic Grouping for Open-Vocabulary 3D Object Detection paper: huggingface.co/papers/2603.21…
00:00