David Fleet รีทวีตแล้ว
We compare our method with the state-of-the-art on both synthetic and experimental benchmarks. Empirically, cryoSPIN outperforms in reconstruction quality and FSC.
[6/7]
English
David Fleet
37 posts
David Fleet
@fleet_dj
Google Research, Brain Team; and University of Toronto.
เข้าร่วม Kasım 2016
18 กำลังติดตาม450 ผู้ติดตาม
David Fleet รีทวีตแล้ว
🎉🎉Excited to share that “CryoSPIN” has been accepted to #NeurIPS2024!
Special thanks to my amazing collaborators and supervisors @DaveLindell, @marcusabrubaker, and @fleet_dj
Paper: arxiv.org/abs/2406.10455
Webpage: shekshaa.github.io/semi-amortized…
Code is also released!
🧵[1/7]
English
David Fleet รีทวีตแล้ว
Why use RLHF when u can backprop through the sampler with any differentiable reward🥹🥹🥹
Paul Vicol@PaulVicol
Check out @clark_kev’s and my paper on fine-tuning diffusion models on differentiable rewards! We present DRaFT, which computes gradients through diffusion sampling. DRaFT is efficient & works across many reward functions. With @kswersk, @fleet_dj arXiv: arxiv.org/abs/2309.17400
English
David Fleet รีทวีตแล้ว
DRaFT backpropagates the reward directly into LoRA parameters – we don’t need to use RL because diffusion sampling is differentiable. We improve efficiency by truncating the BPTT; even truncating to one step still works! (2/5)
English
David Fleet รีทวีตแล้ว
Our best variant, DRaFT-LV, learns 2x faster than ReFL (arxiv.org/abs/2304.05977) and 100x faster than RL. (3/5)
English
David Fleet รีทวีตแล้ว
.@clark_kev & I are excited to share our new work on studying Imagen by evaluating it as a zero-shot classifier! Highlights include Imagen achieving SoTA on Stylized Imagenet and being able to perform attribute binding in certain settings unlike CLIP
arxiv.org/abs/2303.15233
🧵👇
English
David Fleet รีทวีตแล้ว
We can reduce the effect of the fine-tuning or mix different reward functions post-training simply by scaling down or mixing LoRAs. (4/5)
English
David Fleet รีทวีตแล้ว
📄 An updated paper describing 3D Flexible Refinement is now out in @naturemethods!
Paper: nature.com/articles/s4159…
It describes further experimental #cryoEM results and the improved 3DFlex method that was released in #CryoSPARC v4.1 ❄️⚡
Tutorial: guide.cryosparc.com/processing-dat…
Structura Biotechnology@structurabio
1/ We’re thrilled to announce that 3D Flexible Refinement, a motion-based deep generative model for continuous heterogeneity in #cryoEM structures, is available today in #CryoSPARC v4.1 Beta! ❄️⚡ Read more about v4.1: cryosparc.com/updates
English
David Fleet รีทวีตแล้ว
Monocular Depth Estimation using Diffusion Models
abs: arxiv.org/abs/2302.14816
project page: depth-gen.github.io
English
David Fleet รีทวีตแล้ว
Awesome work by Hshmat. The work showcases yet another evidence of how effective noise conditioning augmentation can be.
Hshmat Sahak@HshmatSahak
Thrilled to announce SR3+, our new model that establishes a new state-of-the-art on diffusion-based super-resolution for images in the wild! arxiv.org/abs/2302.07864 Joint work w/ @watson_nn @Chitwan_Saharia @fleet_dj
English
David Fleet รีทวีตแล้ว
Thrilled to announce SR3+, our new model that establishes a new state-of-the-art on diffusion-based super-resolution for images in the wild!
arxiv.org/abs/2302.07864
Joint work w/ @watson_nn @Chitwan_Saharia @fleet_dj
English
David Fleet รีทวีตแล้ว
We've just released the first version of our Deep Learning Tuning Playbook! This is our attempt to distill our process for actually getting good results with deep learning. We emphasize hyperparameter tuning since it has been a large pain point. github.com/google-researc…
English
David Fleet รีทวีตแล้ว
Excited to announce our work on novel view synthesis with diffusion models! Our model can lift a single 2d image into 3d.
3d-diffusion.github.io
Joint work w/ @wchan212 @rmbrualla @hojonathanho @taiyasaki @mo_norouzi
English
David Fleet รีทวีตแล้ว
Excited to announce Imagen Video, our new text-conditioned video diffusion model that generates 1280x768 24fps HD videos! #ImagenVideo
imagen.research.google/video/
Work w/ @wchan212 @Chitwan_Saharia @jaywhang_ @RuiqiGao @agritsenko @dpkingma @poolio @mo_norouzi @fleet_dj @TimSalimans
English
David Fleet รีทวีตแล้ว
Happy to announce DreamFusion, our new method for Text-to-3D!
dreamfusion3d.github.io
We optimize a NeRF from scratch using a pretrained text-to-image diffusion model. No 3D data needed!
Joint work w/ the incredible team of @BenMildenhall @ajayj_ @jon_barron
#dreamfusion
English
David Fleet รีทวีตแล้ว
If you don’t think DallE-2 and Imagen are an Alexnet level moment in the machine learning world you aren’t paying attention enough. Very impressive visual results coming out of these. Getting similar chills to when I saw first web browser, iPhone, etc.
hardmaru@hardmaru
“Oriental painting of tigers wearing VR headsets during the Song dynasty” generated using #Imagen
English
David Fleet รีทวีตแล้ว
Daniel has created some of the most amazing #Imagen images so far. Keep an eye out.
Daniel Smith@motionphi
"A photo of a giant sloth drinking a cup of coffee in the dawn light" Generated with #Imagen, a new text-to-image diffusion model from the very smart folks at Google Brain. I'll be sharing much more, so stay tuned!
English
David Fleet รีทวีตแล้ว
If you have any kind of feedback for the #imagen team, we'd love to hear it! imagen.research.google
@Chitwan_Saharia, @wchan212, @srbhsxn, Lala Li, @jaywhang_, @cephaloponderer, @coolboi95, Burcu Karagol Ayan, Sara Mahdavi, @iraphas13, @TimSalimans, @hojonathanho, @fleet_dj.
English
David Fleet รีทวีตแล้ว
It was 16 years ago, in 2006, that @geoffreyhinton et al released their demo of deep belief nets. Undergrad me was highly impressed, and helped convince me that deep learning was the way to go. I refreshed Geoff's website almost every day checking for new papers... (1/n)
English
David Fleet รีทวีตแล้ว
English