1. X
  2. Ankit Goyal
Log inSign up
Ankit Goyal
223 posts
Image
user avatar
Ankit Goyal
@imankitgoyal
Incoming Assistant Professor @UMRobotics Senior Research Scientist @NVIDIA Foundation Models for Robots, Dexterous Manipulation ex-@Princeton, ex-@IITKanpur
Seattle, WA
imankgoyal.github.io
Joined March 2020
696
Following
3,291
Followers
RepliesRepliesMediaMedia

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
  • Pinned
    user avatar
    Ankit Goyal
    @imankitgoyal
    Jul 2
    Thrilled to share that I'll be joining the University of Michigan as an Assistant Professor in Fall 2027! My lab will work on Robot Learning, Dexterous Manipulation, Robot Foundation Models & 3D Perception. I'm looking for students to join me. Please apply and reach out.
    Image
    Image
    28K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Jun 27, 2023
    Transformer-1D has become the default backbone for NLP and ViT for CV. What about robotics? Introducing RVT: a backbone for object manipulation. Training 36X faster than SOTA PerAct, it learns multiple real-world tasks with only 10 demos/task. Page: robotic-view-transformer.github.io 🧵👇
    Image
    GIF
    62K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Oct 26, 2023
    Applications are open for PhD Research Intern in Robotics NVIDIA for Summer 2024. We are based in Seattle+Bay Area, led by Prof. Dieter Fox. Please consider applying! Appication: nvidia.wd5.myworkdayjobs.com/NVIDIAExternal… More info about our team can be found here: research.nvidia.com/labs/srl/
    27K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Sep 2, 2023
    Applications are now open for Research Scientist in Robotics @nvidia. We are based in Seattle and Bay Area and are led by Prof. Dieter Fox. Please consider applying! nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAEx…
    29K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Sep 18, 2023
    Applications are open for Research Engineer in Robotics NVIDIA. We are based in Seattle and Bay Area and are led by Prof. Dieter Fox. Please consider applying! Appication Link: nvidia.wd5.myworkdayjobs.com/NVIDIAExternal… More info about our team can be found here: research.nvidia.com/labs/srl/
    30K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Aug 15, 2022
    Happy to share that I defended my Ph.D. thesis today @PrincetonCS 🍾👨‍🎓 Next, I will be working as a Research Scientist at @NVIDIAAI with Prof. Dieter Fox. Excited and looking forward to learning about and training robots! 🤖
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Jul 23, 2022
    Finally, read the paper `Masked Autoencoders Are Scalable Vision Learners' (arxiv.org/abs/2111.06377) by Kaiming He et al @MetaAI and was blown away by the results! Here’s my summary and why they are big deal. (1/11)
    Image
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Oct 18, 2021
    Our work on ‘Non-deep Networks’ is out on arxiv. Thanks to the awesome co-authors: @alexeyab84, @jiadeng and Vladlen Koltun.
    user avatar
    AK
    @_akhaliq
    Oct 18, 2021
    Non-deep Networks abs: arxiv.org/abs/2110.07641 By utilizing parallel substructures, a network with a depth of just 12 can achieve top-1 accuracy over 80% on ImageNet, 96% on CIFAR10, and 81% on CIFAR100
    Image
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Nov 4, 2023
    Super excited to attend #CoRL2023! Looking forward to connecting with the robotics community, would love to chat if you are around 😀 We are presenting: RVT: github.com/nvlabs/rvt Shelving, Stacking, Hanging: github.com/anthonysimeono…
    user avatar
    Ankit Goyal
    @imankitgoyal
    Jun 27, 2023
    Transformer-1D has become the default backbone for NLP and ViT for CV. What about robotics? Introducing RVT: a backbone for object manipulation. Training 36X faster than SOTA PerAct, it learns multiple real-world tasks with only 10 demos/task. Page: robotic-view-transformer.github.io 🧵👇
    Image
    GIF
    20K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Aug 30, 2023
    Happy to share that RVT got accepted in CoRL2023 as an Oral! Please checkout the work here🙂: github.com/NVlabs/RVT
    user avatar
    Ankit Goyal
    @imankitgoyal
    Jun 27, 2023
    Transformer-1D has become the default backbone for NLP and ViT for CV. What about robotics? Introducing RVT: a backbone for object manipulation. Training 36X faster than SOTA PerAct, it learns multiple real-world tasks with only 10 demos/task. Page: robotic-view-transformer.github.io 🧵👇
    Image
    GIF
    4.3K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Sep 26, 2022
    Super excited to share our work ProgPromt! We show how LLMs can be used for situated robot planning by prompting them with pythonic code. abs: arxiv.org/abs/2209.11302 project page: progprompt.github.io [1/9]
    user avatar
    AK
    @_akhaliq
    Sep 26, 2022
    ProgPrompt: Generating Situated Robot Task Plans using Large Language Models abs: arxiv.org/abs/2209.11302 project page: progprompt.github.io
    Image
    00:00
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Jul 11, 2023
    Robotic rearrangements tasks shelving books, stacking cans and hanging mugs are multimodal, making them difficult to learn. For this, we propose RPDiff that leverages diffusion to learn pose distribution. Code: github.com/anthonysimeono… 👇 the 🧵 from lead auth. @anthonysimeono_
    This post is unavailable.
    5.5K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Jul 11, 2023
    Thanks @_akhaliq for highlighting our work! We propose a pose diffusion model for 6dof multimodal rearrangement. This work lead by awesome @anthonysimeono_ ! Also thanks to all collaborators and mentors. @lucas_manuelli, @yen_chen_lin, Alina, Alberto, @pulkitology and Dieter.
    user avatar
    AK
    @_akhaliq
    Jul 11, 2023
    Shelving, Stacking, Hanging: Relational Pose Diffusion for Multi-modal Rearrangement paper page: huggingface.co/papers/2307.04… propose a system for rearranging objects in a scene to achieve a desired object-scene placing relationship, such as a book inserted in an open slot of a
    Image
    00:00
    18K
  • user avatar
    Ankit Goyal
    @imankitgoyal
    Aug 12, 2023
    Applications are now open for Nvidia Graduate Fellowship 2024. Please consider applying. Deadline Sep 7 blogs.nvidia.com/blog/2023/08/0…
    5.2K
Advertisement
Advertisement