1. X
  2. Yehonathan Litman
Log inSign up
Yehonathan Litman
172 posts
user avatar

Yehonathan Litman

@yehonation
Ph.D. @CMU_Robotics | AI Researcher @AIatMeta, @RealityLabs
Pittsburgh, PA
yehonathanlitman.github.io
Joined June 2023
193
Following
417
Followers
RepliesRepliesMediaMedia
  • Pinned
    user avatar
    Yehonathan Litman
    @yehonation
    Jun 23
    To be honest, training on handmade 4D asset datasets is a dead-end. Almost all 4D asset data is synthetic and diverse real data barely exists, so models trained on it struggle to reconstruct objects that deform, get occluded, and move freely about the scene. Our new work,
    Image
    00:00
  • user avatar
    Yehonathan Litman
    @yehonation
    Jul 30
    📢 Lift4D (SIGGRAPH Asia '26) Code Release 📢 Generate and reconstruct complete dynamic 4D assets from a single in-the-wild video. 💻github.com/yehonathanlitm… 🔗Webpage: lift4d.github.io 📜Paper: arxiv.org/abs/2606.23688
    Image
    00:00
  • user avatar
    Yehonathan Litman
    @yehonation
    Jul 17
    Diffusion relighting models can produce beautiful relit clips from a text prompt or environment map, but the moment you need real physical control like moving a light, inserting an object, editing materials, or going past 5 seconds, you're helpless. Our new work, LightCrafter,
    Image
    00:00
  • user avatar
    Yehonathan Litman
    @yehonation
    Jun 23
    Thanks for tweeting about our work AK!
    user avatar
    AK
    @_akhaliq
    Jun 23
    Lift4D Harmonizing Single-View 3D Estimation for 4D Reconstruction In-the-Wild
    Image
    00:00
  • user avatar
    Yehonathan Litman
    @yehonation
    May 31
    📢 Code + weights drop for our #CVPR2026 paper! 💻 github.com/yehonathanlitm… 📍Come see our poster presentation #167 Hall A/F on Friday, June 5, 4:00 PM – 6:00 PM
    Image
    Image
    00:12
    user avatar
    Yehonathan Litman
    @yehonation
    Feb 17
    Excited to share our new work EditCtrl! We introduce a disentangled local-global control video inpainting framework that dynamically allocates compute where needed - achieving up to 10x compute savings over full-attention while matching or exceeding SOTA editing quality. 🧵

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Advertisement
Advertisement