Jieneng Chen

Pronounce: jee-eh-nung  ·  陈杰能  ·  Email: jienengc [at] stanford.edu

I am a postdoctoral researcher at Stanford SVL, working with Jiajun Wu. I received my PhD in Computer Science at Johns Hopkins University in 2026, advised by Alan Yuille.

I study intelligence in the physical world by structuring raw observations into spatial code. My works laid the foundation, spanning vision encoders and world models. I am best known for TransUNet, which combines local and global context and is widely recognized as one of the first Transformers for biomedical image analysis. It has received over 10,000 citations.

I am honored as a Young American Scientist by the pretigious magazine Scientific American, a Siebel Scholar in bioengineering, and a recipient of a few best paper awards.

Email CV Scholar Github LinkedIn

News

• 07/2026. I am deeply honored to be recognized by Scientific American as one of the Young American Scientists. The magazine has inspired my curiosity about science since childhood.
• 07/2026. Invited to serve as Area Chair for ICLR 2027 and AAAI 2027.
• 06/2026. Co-organized WMAS (World Model Meets Active Sensing) and GCV (Generative Computer Vision) workshops at IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR) 2026.
• 04/2026. World-in-World is selected as Oral presentation (top 1% of submissions) in ICLR 2026.
• Happy to chat if you’re looking for research collaboration or research opportunities.

Research

When we watch a ball bounce, we do more than recognize pixels. We infer 3D structure, anticipate what may happen next, and reason about physical interactions over time. What representation could give machines the same ability to understand and act in a dynamic world?

I seek fundamental representations of the physical-world observations. My work builds increasingly rich world priors through learned multimodal representations (Chen ’21, ’24, ’26) and generative representations (Lu ’25; Chen ’26). This research path aims to enable machines to reason, plan and adapt through closed-loop interaction (Zhang ’26).

Beyond representations, I apply these ideas to robots operating in complex environments and to pancreatic/liver cancer diagnosis and treatment, with the long-term goal of improving the lives of patients and their families.

• 2026 inaugural class of 28 Young American Scientists, Scientific American

Selected publications. Full list on Google Scholar.

Research Highlights

ICLR'26 Oral

World-in-World

Closed-loop world model — generation, perception & action in the physical world.

NeurIPS'26

CausalSpatial

Causal reasoning in the physical world.

ICLR'25

GenEx

Turn a single image into an explorable 3D world. Agents navigate generated environments.

ICCV'25

MeWM

Medical World Model — generative tumor evolution simulation for personalised treatment planning.

ICLR'26 Oral

World-in-World

Closed-loop world model — generation, perception & action in the physical world.

ICLR '25

GenEx

Turn a single image into an explorable 3D world. Agents navigate generated environments.

ICCV '25

MeWM

Medical World Model — generative tumor evolution simulation for personalised treatment planning.

	World-in-World: World Models in a Closed-Loop World Jiahan Zhang*, Muqing Jiang*, Nanru Dai, TaiMing Lu, Arda Uzunoglu, Shunchi Zhang, Yana Wei, Jiahao Wang, Vishal Patel, Paul Liang, Daniel Khashabi, Cheng Peng, Rama Chellappa, Tianmin Shu, Alan Yuille, Yilun Du, Jieneng Chen. ICLR, 2026. Oral (top 1%). World models live and die by their closed-loop success, not flawless generated visuals. Paper | OpenReview | Project | Leaderboard | Demo | Code
	CausalSpatial: A Comprehensive Benchmark for Object-Centric Causal Spatial Reasoning Wenxin Ma*, Chenlong Wang*, Ruisheng Yuan*, Hao Chen, Nanru Dai, Yijun Yang, Qi Chen, Zhao-Yang Wang, S. Kevin Zhou, Jianwen Xie, Alan Yuille, Jieneng Chen. NeurIPS, Evaluations & Datasets (E&D) Track, 2026. Paper | HuggingFace Dataset
	Fast Generative DeOcclusion for Visual Geometry and Robotics Jieneng Chen*, Tiezheng Zhang*, Xiwei Xuan, Ju He, Yifan Yin, Haojun Shi, Suyu Ye, Xinyi Li, Ruisheng Yuan, Tianmin Shu, Alan Yuille. CVPR, Findings, 2026
	GenEx: Generating an Explorable World TaiMing Lu, Tianmin Shu, Alan Yuille, Daniel Khashabi, Jieneng Chen. ICLR, 2025. Turn a single image into a 3D world adventure. Embodied agents refine beliefs by predicting unseen parts of the physical world. JHU News | Paper | Blog | Project | Code
	Medical World Model: Generative Simulation of Tumor Evolution for Treatment Planning Yijun Yang, Zhao-Yang Wang, Qiuping Liu, Shuwen Sun, Kang Wang, Rama Chellappa, Zongwei Zhou, Alan Yuille, Lei Zhu, Yu-Dong Zhang, Jieneng Chen. ICCV, 2025. Precision medicine via generative world modeling. Paper | Code | Project
	Spatial457: A Diagnostic Benchmark for 6D Spatial Reasoning of Large Multimodal Models Xingrui Wang, Wufei Ma, Tiezheng Zhang, Celso Miguel de Melo, Jieneng Chen†, Alan Yuille†. CVPR, 2025. Highlight. Paper | Code | HuggingFace
	SpatialLLM: A Compound 3D-Informed Design towards Spatially-Intelligent Large Multimodal Models Wufei Ma, Luoxin Ye, Nessa McWeeney, Celso Miguel de Melo, Alan Yuille, Jieneng Chen. CVPR, 2025. Highlight. Paper
	ViTamin: Designing Scalable Vision Models in the Vision-Language Era Jieneng Chen, Qihang Yu, Xiaohui Shen, Alan Yuille, Liang-Chieh Chen. CVPR, 2024. First vision-centric encoder design for LMMs; SoTA on 60+ benchmarks in 2024. Paper | Code | HuggingFace | timm | open_clip
	LLaVolta: Efficient Large Multi-modal Models via Visual Context Compression Jieneng Chen, Luoxin Ye, Ju He, Zhaoyang Wang, Daniel Khashabi, Alan Yuille. NeurIPS, 2024. Paper | Code | Project
	TransUNet: Rethinking the U-Net Architecture Design for Medical Image Segmentation through the Lens of Transformers Jieneng Chen, Jieru Mei, Xianhang Li, Yongyi Lu, Qihang Yu, Qingyue Wei, Xiangde Luo, Yutong Xie, Ehsan Adeli, Yan Wang, Matthew P Lungren, Shaoting Zhang, Lei Xing, Le Lu, Alan Yuille, Yuyin Zhou. Medical Image Analysis, Oct 2024. Top-15 cited 2021 paper in all AI fields. Most downloaded on ScienceDirect. Most cited in MedIA. ICML-W 2021 | Journal | Code

• Invited talk at NSF IAIFI on physics & AI, Boston.
• Lab seminar at Stanford.
• Vision seminar at UIUC.
• Guest lecture at Rice University.
• Lab seminars at Harvard / MIT / HMS / MGH / Dana-Farber.
• Talk at ICLR 2025 Workshop on Embodied Intelligence with LLMs in Open City Environment (slides).
• Talks at JHU: ChemBE, Cognitive Science, CLSP, MINDS, AIEM.

• Instructor: Machine Imagination (EN.601.208), JHU, 2025 & 2026.

• Area Chair: AAAI 2027, ICLR 2027.
• Reviewer: AAAI, CogSci, CVPR, ECCV, ICCV, ICLR, ICML (Reviewer Award), MedIA, MICCAI, Nature Machine Intelligence, NeurIPS, TAPMI, TMI
• Co-organizer: CVPR-GCV 2025/2026, CVPR-WMAS 2026, CVPR-CV4Clinical 2026, ICCV-ESR 2025.
• Inclusion: Lecture for JHU WSE Pre-College Program 2025. JHU CS mentor hours.

• I have been fortunate to mentor and collaborate with many talented undergraduate and master students whose first-author papers have appeared at leading conferences, including ICLR (Lu et al.; Zhang et al.), NeurIPS (Ma et al.), and ICCV (Yang et al.). Their accomplishments also include an Oral recognition (Zhang), a Muuss research award (Lu), and a CRA Outstanding Undergraduate Researcher Award finalist (Lu). Several have since joined PhD programs at Princeton, CMU, Johns Hopkins, UT Austin, Oxford VGG, and Northwestern.