Log inSign up
Oier Mees
347 posts
user avatar
Oier Mees
@oier_mees
Robot Learning & Foundation Models @microsoft & External Lecturer @ETH. Prev. postdoc at @berkeley_ai. PhD @UniFreiburg. Prev. intern @NVIDIAAI.
oiermees.com
Joined October 2022
340
Following
4,962
Followers
  • Pinned
    user avatar
    Oier Mees
    @oier_mees
    Jun 10
    A few years ago, learning robot learning meant stitching together dozens of papers and courses β€” with no clear path from the basics to what state-of-the-art systems actually do. This was one of the motivations behind creating @ETH's course "Robot Learning: From Fundamentals to
    Image
    37K
  • user avatar
    Oier Mees
    @oier_mees
    Sep 19, 2025
    After two fantastic years at @UCBerkeley I'm thrilled to share that I've joined @Microsoft in ZurichπŸ‡¨πŸ‡­to pioneer the next generation of multimodal foundation models to drive agents πŸ€– that can seamlessly interact across the digital and physical worlds 🌍 We are hiring! 🧡
    Image
    59K
  • user avatar
    Oier Mees
    @oier_mees
    Dec 14, 2023
    Thrilled to announce Octo πŸ™, an open-source robot foundation model! Octo is a sota generalist robot policy based on transformer+diffusion. Most importantly, you can finetune Octo *today* with flexible observation and action spaces on your robot setup! octo-models.github.io
    Image
    00:00
    46K
  • user avatar
    Oier Mees
    @oier_mees
    Aug 21, 2024
    We don't have much evidence for that statement. Look at the famous videos from the 2007 Stanford PR1, they already showed impressive skills in unstructured environments, but with teleop. Progress towards generalist robots has clearly been driven by better robot learning software.
    Image
    00:00
    Image
    00:05
    user avatar
    Brett Adcock
    Figure
    @adcock_brett
    Aug 20, 2024
    It's now clear that truly capable hardware has been the key obstacle to advancing general-purpose robotics To build the best AI company in the long run, there is no substitute for good hardware...it's a critical dependency From day one, Figure went all-in on owning our
    20K
  • user avatar
    Oier Mees
    @oier_mees
    Jan 13, 2025
    Can generalist robot πŸ€– policies understand heterogeneous multimodal sensors? Introducing FuSe, a novel finetuning recipe allowing VLAs to reason across πŸ‘οΈ vision, 🀏touch, πŸ”Š sound & πŸ—£οΈ language, enabling queries "pick the object that feels like🍍 but is brown" (hint: pinecone)
    Image
    00:00
    16K
  • user avatar
    Oier Mees
    @oier_mees
    Sep 1, 2023
    Excited to start my Postdoc with Prof. @svlevine at @UCBerkeley! I look forward to collaborating with all the amazing researchers at @berkeley_ai (BAIR)!
    Image
    Image
    6.5K
  • user avatar
    Oier Mees
    @oier_mees
    Oct 16, 2024
    Introducing Diffusion Transformer Block Policies, a sota architecture that outperforms ACT and Diffusion Policies by 20%, needs no hyperparam tuning, scales with model size and data, and enables dexterous long horizon bimanual tasks like sushi cutting! dit-policy.github.io
    Image
    00:00
    8K
  • user avatar
    Oier Mees
    @oier_mees
    Jun 16, 2025
    Can robots build multimodal spatial memories from sight, sound, text enabling crossmodal tasks eg "fetch a coke, avoiding areas where you heard glass breaking" or "find the object I dropped yesterday, I heard a church bell after it"? Check our IJRR paper mslmaps.github.io
    Image
    00:00
    6.1K
  • user avatar
    Oier Mees
    @oier_mees
    Oct 19, 2023
    Today I had the pleasure of presenting my πŸ€– robot learning research at @ToyotaResearch in Los Altos and to learn about their ongoing efforts on building and scaling πŸ“ˆ AI models for robotics! Thanks for the invitation!
    Image
    12K
  • user avatar
    Oier Mees
    @oier_mees
    Mar 14, 2025
    Thrilled to present my research on robot foundation models today in-person at @ETH as part of the Robotics, Vision, and Controls Talks! Talk starts at 16:00 CET; ETH HG E 41, zoom link: robotics-talks.com Come talk to me if you are excited about robot foundation models!
    Image
    00:00
    5.7K
  • user avatar
    Oier Mees
    @oier_mees
    Sep 11, 2023
    Excited that AVLMaps, our take for πŸ€– following multimodal prompts from audio πŸ”Š, vision πŸ‘€ and languageπŸ—£οΈ w/ foundation models to solve zero-shot spatial goal Navigation will be presented at ISER 23! w/ @huang_chenguang @andyzeng_ @wolfram_burgard: avlmaps.github.io
    Image
    00:00
    16K
  • user avatar
    Oier Mees
    @oier_mees
    Oct 22, 2025
    If you have been impacted by today's layoffs at Meta's AI teams, please know that Microsoft Zurich is hiring for positions in multimodal foundation models & robot learning. You and your teams have given so much to the AI community, I hope we can all give back and support you now
    8.2K
  • user avatar
    Oier Mees
    @oier_mees
    Aug 22, 2024
    A truly generalist policy should learn from any robot embodiment. Introducing CrossFormer: a single policy that controls vastly different robots, from drones to quadrupeds, navigators to bimanual manipulators! -🦾SOTA cross-embodied generalist policy - Trained on 900K robot trajs
    Image
    00:00
    9.7K
  • user avatar
    Oier Mees
    @oier_mees
    Nov 5, 2024
    I'm in Munich for #CoRL2024 with new papers about foundation models for and from robotics! Also thrilled to share that I'm on the job market, looking for tenure-track & industry research positions focused on robot learning and embodied AI. I’d love to chat about potential roles!
    Image
    00:00
    6.5K

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.

TermsΒ·PrivacyΒ·CookiesΒ·AccessibilityΒ·Ads InfoΒ·Β© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
Advertisement
Advertisement