Log inSign up
Log inSign up
Anjiang Wei @ COLM 2026
103 posts
@anjiangw

Anjiang Wei @ COLM 2026

@anjiangw
RS @GoogleDeepMind | CS PhD @stanford | BS @PKU1898
CA
cs.stanford.edu/~anjiang/
Joined January 2022
363 Following
412 Followers
RepliesRepliesRepostsRepostsMediaMedia

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·US TIDA·Ads Info·© 2026 X Corp.
  • @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Oct 6
    My collaborators @JOYSUN200402, @TarunSures41845, and I will be presenting Quokka at COLM 2026. 🗓️ Wednesday, Oct. 7, 11:00 AM - 1:00 PM 📍 Imperial Ballroom, Poster #55 Come chat about LLMs for program verification! colm.cc/virtual/2026/p…
    @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Jul 16
    Excited to share that our paper, “Quokka: Accelerating Program Verification with LLMs via Invariant Synthesis,” has been accepted to #COLM2026! 🎉 Paper: arxiv.org/pdf/2509.21629 Code: github.com/Anjiang-Wei/Qu…
  • @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Oct 6
    My collaborator @TarunSures41845 and I will present SuperCoder at COLM 2026! 🗓️ Tuesday, Oct. 6, 11:00 AM - 1:00 PM 📍 Grand Ballroom, Poster #129 Come chat about code RL and making programs faster! colm.cc/virtual/2026/p…
    1
  • @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Sep 30
    Excited to see the progress on cyber and coding!
    @sundarpichai
    Sundar Pichai
    Google
    @sundarpichai
    Sep 30
    Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon! It shows frontier performance in complex workflows, cyber defense and software engineering. Teams are using it extensively at Google, from
    2
  • @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Sep 25
    Check it out!
    @CongyueD
    Congyue Deng
    @CongyueD
    Sep 23
    In my spare time, I wrote a TPU library translating a majority of video models from PyTorch to Jax. 💿Source code: github.com/FlyingGiraffe/… 📑Tech report: arxiv.org/abs/2609.18077 🧊Quick try with PyPI: pypi.org/project/vidax/ Great thanks to @Google TPU Research Cloud (TRC)!
    00:00
  • @anjiangw
    Anjiang Wei @ COLM 2026
    @anjiangw
    Sep 18
    Awesome work!
    @TarunSures41845
    Tarun Suresh @ COLM 2026
    @TarunSures41845
    Sep 18
    Diffusion LLMs and speculative decoding promise much faster agents. Yet agents need long contexts, and training on them is painfully slow. Introducing Context-Sharded Block Parallelism (CSBP), a new distributed parallelism strategy unlocking significant training efficiency for