Ji Xie

I am a senior undergraduate student at Zhejiang University, pursuing a Bachelor of Engineering in Computer Science and Technology with Honors from Chu Kochen Honors College. I expect to graduate in June 2026. Rank: 2/147. GPA: 93.8/100.

Currently, I am a research intern at the Berkeley AI Research (BAIR) Lab, UC Berkeley, advised by Prof. XuDong Wang and Prof. Trevor Darrell.

My long-term goal is to build a unified, controllable, and powerful multimodal model and apply its strong generative priors to the construction of world models and embodied AI.

Email  /  Google Scholar  /  GitHub  /  Twitter

profile photo

Selected Research

I'm interested in computer vision, generative models, and multimodal learning.

reca Reconstruction Alignment Improves Unified Multimodal Models
Ji Xie, Trevor Darrell, Luke Zettlemoyer, Xudong Wang
ICLR 2026
paper / code / model

A single reconstruction loss improves generation and editing capabilities of Unified Models.

icedit In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large-Scale Diffusion Transformer
Zechuan Zhang, Ji Xie, Yu Lu, Zongxin Yang, Yi Yang
NeurIPS 2025
paper / code (2K Stars🌟) / model

ICEdit enables instructional image editing through in-context generation in large-scale diffusion transformers.

3dis 3DIS: Depth-Driven Decoupled Instance Synthesis for Text-to-Image Generation
Dewei Zhou*, Ji Xie*, Zongxin Yang, Yi Yang
ICLR 2025 (Spotlight)
paper / code / model

3DIS uses depth-driven decoupled instance synthesis for controllable text-to-image generation.

Invited Talks

"Reconstruction Alignment Improves Unified Multimodal Model"
Apple Research · Invited Talk · Hosted by Chen Chen and Yinfei Yang
October 2025

Selected Honors & Awards

SenseTime Scholarship
Top 30 recipients annually in China
June 2025
Gold Medal, International Collegiate Programming Contest (ICPC), Regional
October 2022
Gold Medal, China Collegiate Programming Contest (CCPC), Regional
October 2022

Miscellaneous

I was a member of the ZJU ACM/ICPC team and achieved a rating of 2478 on Codeforces. You can find my old blog here — it contains my competitive-programming notes :)


Website template from Jon Barron