Log inSign up
Image
user avatar
Tim Salimans
@TimSalimans
AI Research Scientist
Amsterdam
Joined March 2013
458
Following
7,963
Followers
  • user avatar
    Tim Salimans
    @TimSalimans
    Oct 5, 2022
    Very happy to release #ImagenVideo today! Amazing work with an amazing team! imagen.research.google/video/ High fidelity text to video with diffusion models: "Flying through an intense battle between pirate ships in a stormy ocean."
    Image
    00:00
    Image
    00:05
    user avatar
    Jonathan Ho
    @hojonathanho
    Oct 5, 2022
    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
  • user avatar
    Tim Salimans
    @TimSalimans
    Feb 4, 2022
    Diffusion models can now produce great samples in as few as 4 sampling steps, as @hojonathanho and I show in our new paper (ICLR spotlight): openreview.net/forum?id=TIdIX… 🧵 1/
    Image
  • user avatar
    Tim Salimans
    @TimSalimans
    Jul 10, 2018
    Looking forward to my first ICML as a Googler, working alongside Lasse Espeholt and @NalKalchbrenner at a new Google Brain office in Amsterdam! #ICML2018 #GoogleBrain
  • user avatar
    Tim Salimans
    @TimSalimans
    Aug 5, 2020
    Our paper on implicit generative models for speech synthesis is out! We use a spectrogram-based loss like DDSP and OpenAI Jukebox, but add a repulsive term that offers statistical guarantees: arxiv.org/abs/2008.01160 w/ @agritsenko @vdbergrianne @NalKalchbrenner @latentjasper 1/3
    Image
  • user avatar
    Tim Salimans
    @TimSalimans
    Jun 6, 2024
    My team in Amsterdam is hiring a research scientist with experience in generative modeling. Please apply if you want to help us build the next generation of Google's generative models!
    job-boards.greenhouse.io
    DeepMind
    57K
  • user avatar
    Tim Salimans
    @TimSalimans
    Jun 11, 2024
    We have a new distillation method that actually *improves* upon its teacher. Moment Matching distillation (arxiv.org/abs/2406.04103) creates fast stochastic samplers by matching data expectations between teacher and student. Work with @emiel_hoogeboom @JonathanHeek @tejmensin. 1/4
    Image
    21K
  • user avatar
    Tim Salimans
    @TimSalimans
    Jul 6, 2021
    Diffusion-based generative models are SOTA on perceptual quality, but how good are they at density estimation? Very good it turns out! 📢Announcing our new paper on diffusion models, with SOTA likelihoods & compression + new theoretical insights into this exciting model class.
    user avatar
    Durk Kingma
    @dpkingma
    Jul 6, 2021
    New paper: Variational Diffusion Models (VDMs)! arxiv.org/abs/2107.00630 ✅ New general insights into diffusion models ✅ Simple objective ✅ Fast optimization & anytime synthesis ✅ SotA likelihoods & lossless compression Work with @TimSalimans @poolio @hojonathanho (1/n)
    Image
  • user avatar
    Tim Salimans
    @TimSalimans
    Jan 27, 2023
    With a simple modification to the noise schedule, diffusion models now work directly at high resolutions without requiring cascades or latents! And it's fast too: 0.4 seconds to generate an image using our distilled model. Check out the paper with @emiel_hoogeboom & @JonathanHeek
    user avatar
    Emiel Hoogeboom
    @emiel_hoogeboom
    Jan 27, 2023
    If diffusion models are so great, why do they require modifications to work well? Like latent diffusion and superres diffusion? Introducing "simple diffusion": a single straightforward diffusion model for high res images (arxiv.org/abs/2301.11093) . w/ @JonathanHeek @TimSalimans
    picture of two robots playing chess
    29K
  • user avatar
    Tim Salimans
    @TimSalimans
    Mar 12, 2024
    New work on distillation of diffusion models 📢 Getting closer to zero loss in quality than ever before, while needing only 4-16 sampling steps. Bonus content: a new deterministic sampler - aDDIM - generalizing DDIM and performing substantially better!
    user avatar
    Jonathan Heek
    @JonathanHeek
    Mar 12, 2024
    Fast sampling with 'Multistep Consistency Models': We get 1.6 FID on Imagenet64 in 4 steps and scale text-to-image models, generating 256x256 images with 16 steps. Guess which row is distilled? With @emiel_hoogeboom @TimSalimans Arxiv: arxiv.org/abs/2403.06807
    Image
    8.1K
  • user avatar
    Tim Salimans
    @TimSalimans
    Oct 8, 2018
    We're now accepting internship applications for 2019! careers.google.com/jobs#!t=jo&jid…
  • user avatar
    Tim Salimans
    @TimSalimans
    Oct 10, 2022
    Progressive distillation now works for classifier-free guided diffusion models and stochastic samplers! Excited to share this work with Chenlin and other collaborators on how to make diffusion models fast to sample.
    user avatar
    Chenlin Meng
    @chenlin_meng
    Oct 10, 2022
    Excited to share our work "On distillation of guided diffusion models"! Our distillation approach allows classifier-free guided diffusion models to generate high-quality samples using as few as 1-4 sampling steps😮
    Image
  • user avatar
    Tim Salimans
    @TimSalimans
    Feb 4, 2022
    Replying to @TimSalimans
    We achieve this by taking a slow model using as many as 8192 steps, and progressively halving the number of steps through distillation. At each iteration, the student learns to amortize 2 steps of the deterministic DDIM teacher sampler into a single step. 2/
    Image
  • user avatar
    Tim Salimans
    @TimSalimans
    May 7, 2024
    Replying to @iclr_conf and @dpkingma
    Congratulations @dpkingma and @wellingmax! They couldn't have made a better choice.
    4.2K
  • user avatar
    Tim Salimans
    @TimSalimans
    Feb 4, 2022
    Replying to @TimSalimans
    Every iteration of distillation warm-starts from the previous one, so this whole process is fast! Even though our distilled models are as small as their original teacher, they show little degradation in sample quality down to as few as 4 sampling steps. 3/3
    Image