Oğuzhan Fatih Kar

@oguzhanthefatih

ML Researcher at @Apple. Previously @EPFL_en, @Google. Into multimodality.

Switzerland

ofkar.github.io

Replies Reposts Media

Pinned
Oğuzhan Fatih Kar
@oguzhanthefatih
May 13, 2025
Happy to share that I’ve successfully defended my PhD thesis, “Scaling the Modalities in Multimodal Foundation Models”! 🎓 🎉 A huge thanks to my incredible advisor @zamir_ar and all the amazing collaborators I’ve had the chance to work with across EPFL, Apple, and Google.
8
Oğuzhan Fatih Kar
@oguzhanthefatih
Jul 9
Interested in test-time search and unified multimodal modeling? Come chat with us at the posters today!
Zhitong Gao
@zhitong_gao
Jul 7
Excited to be at #ICML this week! I'll be presenting our works: 🔹 (1D) Ordered Tokens Enable Efficient Test-Time Search (soto.epfl.ch) on Thu, Jul 9 · 5:00–6:45 PM · Hall A #2504 🔹 MODUS: Decoder-only Any-to-Any Modeling of Diverse Modalities
Oğuzhan Fatih Kar
@oguzhanthefatih
Jul 8
We present VideoFlexTok this afternoon, come chat with us at the poster!
Andrei Atanov
@andrew_atanov
Jul 5
Flying to Seoul to present our #ICML2026 Spotlight VideoFlexTok! Poster Session 4 (14:30-16:15) on Wednesday, poster 3001. Come say hi! Also don’t miss the expo talk by @oguzhanthefatih on Monday on (Video)FlexTok!
Oğuzhan Fatih Kar
@oguzhanthefatih
Jul 4
Heading to ICML 2026 in Seoul! 🇰🇷 🎙️I’ll be giving an Expo Talk on Monday on VideoFlexTok (Spotlight), our work on flexible-length, coarse-to-fine video tokenization: videoflextok.epfl.ch We’ll also present the following papers: • (1D) Ordered Tokens Enable Efficient
00:00
Oğuzhan Fatih Kar
@oguzhanthefatih
Jun 2
Interested in multimodal research? Consider applying to our team in Zürich🇨🇭 And catch Roman at CVPR to learn more!
Roman Bachmann
@roman__bachmann
Jun 2
We're hiring for MINT-Zürich 🇨🇭 @Apple , interns + full-time to build multimodal foundation models, with research across pre-training, post-training, agents, and world models! Email us at [email protected] with CV + relevant work, and apply to