Log inSign up
Junting Pan
140 posts
Image
user avatar
Junting Pan
@junting9
Research Scientist @ Apple AIML | Prev: PhD@MMLab CUHK, Intern @AIatMeta (FAIR) and @samsungresearch. Working on foundation models.
junting.github.io
Joined December 2013
467
Following
990
Followers
  • Pinned
    user avatar
    Junting Pan
    @junting9
    Oct 15, 2024
    Our MathVision benchmark is accepted by NeurIPS DB Track, 2024! We show a notable performance gap between current LMMs and human performance on simple math problems with visual context. Dataset: huggingface.co/datasets/MathL… Paper: arxiv.org/pdf/2402.14804
    Image
    Image
    4.2K
  • user avatar
    Junting Pan
    @junting9
    Oct 14, 2025
    The Foundation Model Team @🍎Apple AI/ML is looking for a Research Intern (flexible start date) to work on Multimodal LLMs and Vision-Language. Interested? DM me to learn more!
    31K
  • user avatar
    Junting Pan
    @junting9
    Oct 29, 2022
    My first in person Conference as a Ph.D student. Glad to meet so many great researchers, hope to see you all soon! #ECCV2022 Next step #NeurIPS2022 !
    Image
    Image
    Image
    Image
  • user avatar
    Junting Pan
    @junting9
    Jul 30, 2024
    SAM2 is truly amazing! I am so proud to have been a part of this incredible team!
    user avatar
    AI at Meta
    Meta
    @AIatMeta
    Jul 29, 2024
    Introducing Meta Segment Anything Model 2 (SAM 2) — the first unified model for real-time, promptable object segmentation in images & videos. SAM 2 is available today under Apache 2.0 so that anyone can use it to build their own experiences Details ➡️ go.fb.me/p749s5
    Image
    00:00
    5.9K
  • user avatar
    Junting Pan
    @junting9
    Oct 29, 2022
    Replying to @junting9
    x.com/_akhaliq/statu… The code is open-sourced at github.com/saic-fi/edgevit!
    user avatar
    AK
    @_akhaliq
    May 10, 2022
    EdgeViTs: Competing Light-weight CNNs on Mobile Devices with Vision Transformers abs: arxiv.org/abs/2205.03436
    Image
  • user avatar
    Junting Pan
    @junting9
    Nov 29, 2022
    Today, I will present our #NeurIPS2022 paper ST-Adapter a super-efficient method that adapts Image Foundation Models for Video Recognition at Hall J #631. at 11am.
    Image
  • user avatar
    Junting Pan
    @junting9
    Oct 23, 2022
    Hello @#ECCV from Tel Aviv!
    Image
  • user avatar
    Junting Pan
    @junting9
    Oct 24, 2022
    I will be presenting our 5 winning solution at @ego4_d workshop. Feel free to approach and chat 😉. #ECCV
    Image
  • user avatar
    Junting Pan
    @junting9
    Nov 27, 2022
    The registration desk is only about 2km from the main entrance. 😂 Keep walking! #NeurIPS2022
    Image
  • user avatar
    Junting Pan
    @junting9
    May 21, 2022
    OMG! 98.24 Top1-Accuracy on Kinetics-400 and 96.35 on Kinetics-700?? 😱How can this happen? Action Recognition -> Solved? @3scorciav @yass_ouali
    Image
    Papers with code
    From github.com
  • user avatar
    Junting Pan
    @junting9
    Nov 30, 2022
    I was the normal sized man and hope the poster was clear enough for everyone 🤣
    user avatar
    Bored Yann LeCun
    @boredyannlecun
    Nov 29, 2022
    Replying to @boredyannlecun
    Normal sized man, abnormally large poster. #feelthelearn #neurips2022
    Image
  • user avatar
    Junting Pan
    @junting9
    Nov 27, 2022
    Hi #NeurIPS2022@ from NOLA!
    Image
  • user avatar
    Junting Pan
    @junting9
    Jun 21, 2021
    Check out our #CVPR2021 work 1583 “Actor-Context-Actor Relation Network for Spatio-temporal Action Localization” (Today 11:00 - 13:00 EDT in Paper Session 1) Paper: arxiv.org/abs/2006.07976 Code & Models: github.com/Siyu-C/ACAR-Net x.com/junting9/statu…
    Image
  • user avatar
    Junting Pan
    @junting9
    Oct 25, 2024
    Nice to e-meet everyone at #AdobeMAX sneak! @mengweir --- Now I know 😎
    user avatar
    Mengwei Ren
    @mengweir
    Oct 16, 2024
    So much fun at #AdobeMAX sneak! As a researcher, this is so far the biggest stage I have ever stepped on. Turns out it’s easier to interact with 10k+ audiences than with 1 for an introvert :p Grateful for all the applause, energy, and suports!
    574

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