Yang Luo

18 posts

Yang Luo

@YangL_7

Ph.D. candidate @NUSingapore | Efficient ML | ACL Outstanding Paper Award

Singapore Beigetreten Aralık 2021

37 Folgt127 Follower

Angehefteter Tweet

Yang Luo@YangL_7·20 Ara

Training-free Video Enhancement: Achieved 🎉 Nice work with @oahzxl @shaowenqi126301 @VictorKaiWang1 @VitaGroupUT @YangYou1991 et al. Non-trivial enhancement, training-free, and plug-and-play 🥳 Blog: oahzxl.github.io/Enhance_A_Vide… (🧵1/6)

English

251

46.4K

Yang Luo@YangL_7·5 Mar

Amazing work on parameter generation!!

Tencent HY@TencentHunyuan

One static model does not fit all😭 We just dropped our latest work: Functional Neural Memory. Instead of static models, we generate custom "parameters" for every single input. ✅Prompt your model anytime ✅Instant personalization ✅Better instruction following ✅Flexible & dynamic memory (w/o memory bank✌️) (🧵1/6)

English

375

Yang Luo@YangL_7·23 Haz

Customize your LLM in seconds!

AK@_akhaliq

Drag-and-Drop LLMs Zero-Shot Prompt-to-Weights

English

165

Yang Luo retweetet

Victor.Kai Wang@VictorKaiWang1·20 Haz

Customizing Your LLMs in seconds using prompts🥳! Excited to share our latest work with @HPCAILab, @VITAGroupUT, @k_schuerholt, @YangYou1991, @mmbronstein, @damianborth : Drag-and-Drop LLMs(DnD). 2 features: tuning-free, comparable or even better than full-shot tuning.(🧵1/8)

English

114

17.7K

Yang Luo retweetet

Ziqiao Wang@ZiqiaoWang63428·27 May

Representation Alignment (REPA) is NOT ALWAYS helpful for diffusion training!🤷 Sharing latest work w/ @HPCAILab and @VITAGroupUT: "REPA Works Until It Doesn't: Early-Stopped, Holistic Alignment Supercharges Diffusion Training". Acceleration up to 28x w/o performance drop.(🧵1/7)

English

3.8K

Yang Luo@YangL_7·28 Şub

Glad to see CAME's efficiency in the training of SANA 1.5!

Enze Xie@xieenze_jr

🔥 SANA 1.5: A linear Diffusion Transformer pushes SOTA in text-to-image generation! Key innovations: • Depth-growth training: 1.6B → 4.8B params • Memory-efficient 8-bit optimizer • Flexible model pruning • Inference scaling for better quality Achieves 0.80 on GenEval! 🚀

English

284

Yang Luo retweetet

Ziming Liu@lzm_mlsys·17 Şub

🚀Towards efficient Diffusion Transformers! 😆We are happy to introduce RAS, the first diffusion sampling strategy that allows for regional variability in sampling ratios, achieving up to 2x+ speedup! 🔌Training-free, plug and play! 💪Nice work with @MSFTResearch @YangYou1991 @Yif_Yang et al. 📜Paper: huggingface.co/papers/2502.10… 📖Blog: aka.ms/ras-dit ⌨️Code: github.com/microsoft/RAS (1/5)

English

188

18K

Yang Luo retweetet

Victor.Kai Wang@VictorKaiWang1·20 Oca

Generating ~200 million parameters in just minutes! 🥳 Excited to share our work with @MTDovent , @heisejiasuo96 , and @YangYou1991: 'Recurrent Diffusion for Large-Scale Parameter Generation' (RPG for short). Example: Obtain customized models using prompts (see below). (🧵1/8)

English

286

45.2K

Yang Luo@YangL_7·22 Ara

@nirzabari @oahzxl @shaowenqi126301 @VictorKaiWang1 @VITAGroupUT @YangYou1991 Thank you! Glad to hear the improvements worked well with LTXV too. Your observation about the enhanced coherence in movement and detail preservation is super helpful feedback.

English

277

Nir Zabari‎@nirzabari·22 Ara

@YangL_7 @oahzxl @shaowenqi126301 @VictorKaiWang1 @VITAGroupUT @YangYou1991 Excellent analysis and improvements! I implemented this for LTXV and noticed enhancements as well. The most notable improvements are the reduction in people/animals/etc splitting into multiple parts (leading to more coherent movement) and finer details. Great work!

English

333

Yang Luo@YangL_7·20 Ara

English

251

46.4K

Yang Luo@YangL_7·22 Ara

@ai_kimera @oahzxl @shaowenqi126301 @VictorKaiWang1 @VITAGroupUT @YangYou1991 Thanks! Really appreciate you noticing the progress, and we're excited to keep pushing forward.

English

177

Ken Kimera@ai_kimera·21 Ara

@YangL_7 @oahzxl @shaowenqi126301 @VictorKaiWang1 @VITAGroupUT @YangYou1991 Congratulations on this. I've seen some real examples before and after and it looks like you guys achieved a huge milestone.

English

215

Yang Luo@YangL_7·20 Ara

For more details, check out our blog: oahzxl.github.io/Enhance_A_Video Our code is open-source on GitHub: github.com/NUS-HPC-AI-Lab… More results and the technique report will be coming soon! (🧵6/6)

English

841

Yang Luo@YangL_7·20 Ara

Our approach: we compute the average of non-diagonal elements of temporal attention maps as cross-frame intensity (CFI). An enhanced temperature parameter multiplies the CFI to enhance the temporal attention output.

English

1.2K

Yang Luo@YangL_7·18 Mar

@oahzxl Lei God

Norsk

Xuanlei Zhao@oahzxl·18 Mar

@YangL_7 No wonder it's ACL 2023 Outstanding Paper Award winner - being able to train impressive DiT models like PIXART-Σ with your CAME optimizer shows the real-world impact of your work!

English

Yang Luo@YangL_7·17 Mar

We are pleased to announce that PIXART-Σ is trained by our CAME optimizer(arxiv.org/abs/2307.02047). Glad to see that our work has a real-world impact on the training of DiT models! Code: github.com/yangluo7/CAME

English

264

Yang Luo retweetet

Zangwei Zheng@ZangweiZheng·15 Tem

Explore the blog post for a concise and insightful overview of the CAME optimizer! Congrats to the first author @YangYoungLL! 🔥 Blog: zhengzangw.github.io/blogs/came ArXiv: arxiv.org/abs/2307.02047 Code: github.com/huawei-noah/Pr… (a plug-and-play optimizer repo will be released soon)

Zangwei Zheng@ZangweiZheng

🎉 We're thrilled and honored to have received the ACL Outstanding Paper Award! 🏆 Stay tuned, as we'll be publishing a repository for the CAME optimizer very soon. #ACL2023

English

985

Entdecken

@HPCAILab @VITAGroupUT @k_schuerholt @YangYou1991 @mmbronstein @damianborth @MSFTResearch @Yif_Yang