Jang Hyun (Vincent) Cho

Research Scientist · Meta Superintelligence Labs

I am on the multimodal pretraining team, building the next generation of Muse Spark.

I received my Ph.D. in Computer Science from UT Austin, advised by Prof. Philipp Krähenbühl. Before that, I received my B.S. from Cornell University, where I worked with Prof. Bharath Hariharan and Prof. Kavita Bala. I have previously worked at Meta FAIR and NVIDIA.

News

Dec 2025Perception Encoder was presented as an oral at NeurIPS 2025, and PerceptionLM as a spotlight.
Apr 2025I completed my Ph.D. defense at UT Austin.
Jan 2025Cube-LLM was accepted to ICLR 2025.
Jan 2025I joined Meta as a Research Scientist.

Publications

All on Scholar ↗

2026

The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes

A. Adcock, A. Srivastava, A. Dubey, et al., including Jang Hyun Cho

Technical report, 2026

scholar

2025

PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding

Jang Hyun Cho*, Andrea Madotto*, Effrosyni Mavroudi*, Triantafyllos Afouras*, Tushar Nagarajan*, Muhammad Maaz*, Yale Song*, Tengyu Ma*, et al.

NeurIPS 2025 Spotlight · *Joint first authors

paper code

Perception Encoder: The Best Visual Embeddings Are Not at the Output of the Network

Daniel Bolya*, Po-Yao Huang*, Peize Sun*, Jang Hyun Cho*, Andrea Madotto*, Chen Wei, Tengyu Ma, Jiale Zhi, et al.

NeurIPS 2025 Oral · *Joint first authors

paper code

VILA²: VILA Augmented VILA

Yunhao Fang*, Ligeng Zhu*, Yao Lu, Yan Wang, Pavlo Molchanov, Jan Kautz, Jang Hyun Cho, Marco Pavone, Song Han, Hongxu Yin

ICCV Workshops 2025 Oral

paper

Language-Image Models with 3D Understanding

Jang Hyun Cho, Boris Ivanovic, Yulong Cao, Edward Schmerling, Yue Wang, Xinshuo Weng, Boyi Li, Yurong You, Philipp Krähenbühl, Yan Wang, Marco Pavone

ICLR 2025

paper project

2024