Scholar
Xiujun Li
Google Scholar ID: SW_WaQ0AAAAJ
University of Washington / Apple
Citations & Impact
Citations
9,696
H-index
37
i10-index
59
Publications
20
Co-authors
48
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Text as Images: Can Multimodal Large Language Models Follow Printed Instructions in Pixels? (arXiv 2024)
- Ferret-UI 2: Mastering Universal User Interface Understanding Across Platforms (ICLR 2025)
- Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset (ICLR 2025)
- Multimodal Autoregressive Pre-training of Large Vision Encoders (CVPR 2025)
- VinVL: Making Visual Representations Matter in Vision-Language Models (CVPR 2021)
- Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks (ECCV 2020)
- Robust navigation with language pretraining and stochastic sampling (EMNLP 2019)
- End-to-End Task-Completion Neural Dialogue Systems (IJCNLP 2017)
- Composite Task-Completion Dialogue System via Hierarchical Deep Reinforcement Learning (EMNLP 2017)
- Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning (ACL 2018)
Research Experience
- Worked five years at Microsoft Research before joining Apple. Research experience spans dialog, deep reinforcement learning, NLP, vision and language, and multimodal LLMs.
Education
- PhD from UW CSE in 2024, advised by Yejin Choi.
Background
- Research interests include multimodal LLMs, LLMs, NLP, vision and language. Currently a Research Scientist at Apple. Also interested in video generation.
