Xiang Li 李希昂

[01/2026] We launched Grok Imagine v1!
[10/2025] We launched Veo3.1, a strong video model!
[08/2025] One paper accepted to EMNLP 2025 (findings).
[05/2025] We launched Veo3 and Veo2 controls, try it out!
[05/2025] One paper accepted to ICML 2025.
[02/2025] One paper accepted to CVPR 2025.
[01/2025] Two papers (one first-author) accepted to ICLR 2025.
[09/2024] Three papers accepted to NeurIPS 2024.
[06/2024] One first-author paper accepted to ECCV 2024.
[06/2024] One paper accepted to InterSpeech 2024.
[05/2024] Two papers (one first-author) accepted to ICML 2024.
[03/2024] One paper accepted to NAACL 2024.
[02/2024] One first-author paper accepted to CVPR 2024.
[12/2023] One paper accepted to ICASSP 2024.
[10/2023] One first-author paper accepted to EMNLP 2023.
[09/2023] One first-author paper accepted to NeurIPS 2023.
[06/2023] One first-author paper accepted to ICCV 2023.

Publications

Grok Imagine Core Contributor Full stack contribution from data, diffusion model and VLM. [Page]

Veo Core Contributor Contribute to I2V, R2V and Video Editing. [Page]

Image Tokenizer Needs Post Training Kai Qiu*, Xiang Li*, Hao Chen, Jason Kuen, Xiaohao Xu, Jiuxiang Gu, Yinyi Luo, Bhiksha Raj, Zhe Lin, Marios Savvides Preprint 2025 [paper] [project page]

CAARMA: Class Augmentation with Adversarial Mixup Regularization Massa Baali, Xiang Li, Hao Chen, Rita Singh, Bhiksha Raj EMNLP 2025 (Findings) [paper] [code]

Robust Latent Matters: Boosting Image Generation with Sampling Error Synthesis Kai Qiu*, Xiang Li*, Jason Kuen, Hao Chen, Xiaohao Xu, Jiuxiang Gu, Yinyi Luo, Bhiksha Raj, Zhe Lin, Marios Savvides preprint [paper] [code]

Towards Ambiguity-Free Spatial Foundation Model: Rethinking and Decoupling Depth Ambiguity Xiaohao Xu, Feng Xue, Xiang Li, Haowei Li, Shusheng Yang, Tianyi Zhang, Matthew Johnson-Roberson, Xiaonan Huang preprint [paper] [code]

Masked Autoencoders Are Effective Tokenizers for Diffusion Models Hao Chen, Yujin Han, Fangyi Chen, Xiang Li, Yidong Wang, Jindong Wang, Ze Wang, Zicheng Liu, Difan Zou, Bhiksha Raj ICML 2025 [paper] [code]

SoftVQ-VAE: Efficient 1-Dimensional Continuous Tokenizer Hao Chen, Ze Wang, Xiang Li, Ximeng Sun, Fangyi Chen, Jiang Liu, Jindong Wang, Bhiksha Raj, Zicheng Liu, Emad Barsoum CVPR 2025 [paper] [code]

XQ-GAN: An Open-source Image Tokenization Framework for Autoregressive Generation Xiang Li*, Kai Qiu*, Hao Chen, Jason Kuen, Jiuxiang Gu, Bhiksha Raj, Zhe Lin preprint [paper] [code]

On the Diversity of Synthetic Data and its Impact on Training Large Language Models Hao Chen, Abdul Waheed, Xiang Li, Yidong Wang, Jindong Wang, Bhiksha Raj, Marah I Abdin preprint [paper]

Efficient Autoregressive Audio Modeling via Next-Scale Prediction Kai Qiu, Xiang Li, Hao Chen, Jie Sun, Jinglu Wang, Zhe Lin, Marios Savvides, Bhiksha Raj preprint [paper] [code]

ControlVAR: Exploring Controllable Visual Autoregressive Modeling Xiang Li, Kai Qiu, Hao Chen, Jason Kuen, Zhe Lin, Rita Singh, Bhiksha Raj preprint [paper] [code]

ImageFolder: Autoregressive Image Generation with Folded Tokens Xiang Li, Hao Chen, Kai Qiu, Jason Kuen, Jiuxiang Gu, Bhiksha Raj, Zhe Lin ICLR 2025 [paper] [code] [project page]

Scalable Benchmarking and Robust Learning for Noise-Free Ego-Motion and 3D Reconstruction from Noisy Video Xiaohao Xu, Tianyi Zhang, Shibo Zhao, Xiang Li, Sibo Wang, Yongqi Chen, Ye Li, Bhiksha Raj, Matthew Johnson-Roberson, Sebastian Scherer, Xiaonan Huang ICLR 2025 [paper]

Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations Hao Chen, Ankit Shah, Jindong Wang, Ran Tao, Yidong Wang, Xiang Li, Xing Xie, Masashi Sugiyama, Rita Singh, Bhiksha Raj NeurIPS 2024 [paper]

Slight Corruption in Pre-training Data Makes Better Diffusion Models Hao Chen, Yujin Han, Diganta Misra, Xiang Li, Kai Hu, Difan Zou, Masashi Sugiyama, Jindong Wang, Bhiksha Raj NeurIPS 2024 [paper]

Efficient LLM Jailbreak via Adaptive Dense-to-sparse Constrained Optimization Kai Hu, Weichen Yu, Tianjun Yao, Xiang Li, Wenhe Liu, Lijun Yu, Yining Li, Kai Chen, Zhiqiang Shen, Matt Fredrikson NeurIPS 2024 [paper]

R^2-Bench: Benchmarking the Robustness of Referring Perception Models under Perturbations Xiang Li, Kai Qiu, Jinglu Wang, Xiaohao Xu, Rita Singh, Kashu Yamazaki, Hao Chen, Xiaonan Huang, Bhiksha Raj ECCV 2024 [paper] [code]

Evaluating and Improving Continual Learning in Spoken Language Understanding Muqiao Yang, Xiang Li, Umberto Cappellazzo, Shinji Watanabe, Bhiksha Raj InterSpeech 2024 [paper]

Completing Visual Objects via Bridging Generation and Segmentation Xiang Li, Yinpeng Chen, Chung-Ching Lin, Rita Singh, Bhiksha Raj, Zicheng Liu ICML 2024 [paper]

Rethinking Audiovisual Segmentation with Semantic Quantization and Decomposition Xiang Li, Jinglu Wang, Xiaohao Xu, Rita Singh, Yan Lu, Bhiksha Raj CVPR 2024 [paper] [code]

AutoPRM: Self-supervised Fine-grained Feedback for Multi-Step Reasoning via Controllable Question Decomposition Zhaorun Chen, Zhuokai Zhao, Zhihong Zhu, Ruiqi Zhang, Xiang Li, Bhiksha Raj, Huaxiu Yao NAACL 2024 (short version at ICLR 2024 R2-FM workshop) [paper]

A General Framework for Learning from Weak Supervision Hao Chen, Jindong Wang, Lei Feng, Xiang Li, Yidong Wang, Xing Xie, Masashi Sugiyama, Rita Singh, Bhiksha Raj ICML 2024 [paper]

Customizable Perturbation Synthesis for Robust SLAM Benchmarking Xiaohao Xu, Tianyi Zhang, Sibo Wang, Xiang Li, Yongqi Chen, Ye Li, Bhiksha Raj, Matthew Johnson-Roberson, Xiaonan Huang preprint [paper] [code]

A Closer Look at Reinforcement Learning-based Automatic Speech Recognition Fan Yang, Muqiao Yang, Xiang Li, Yuxuan Wu, Zhiyuan Zhao, Bhiksha Raj, Rita Singh Computer Speech & Language [paper]

Improving Continual Learning of Acoustic Scene Classification via Mutual Information Optimization Muqiao Yang, Umberto Cappellazzo, Xiang Li, Shinji Watanabe, Bhiksha Raj ICASSP 2024 [paper]

Towards Noise-Tolerant Speech-Referring Video Object Segmentation: Bridging Speech and Text Xiang Li, Jinglu Wang, Xiaohao Xu, Muqiao Yang, Rita Singh, Bhiksha Raj EMNLP 2023 [paper]

PaintSeg: Training-free Segmentation via Painting Xiang Li, Chung-Ching Lin, Yinpeng Chen, Jinglu Wang, Zicheng Liu, Bhiksha Raj NeurIPS 2023 [paper] [code]

Rethinking Voice-Face Correlation: A Geometry View Xiang Li, Yandong Wen, Muqiao Yang, Jinglu Wang, Rita Singh, Bhiksha Raj ACM Multimedia, 2023 [paper] [code]

Robust Referring Video Object Segmentation with Cyclic Structural Consensus Xiang Li, Jinglu Wang, Xiaohao Xu, Xiao Li, Yan Lu, Bhiksha Raj ICCV, 2023 [paper] [project page] [code]

The Hidden Dance of Phonemes and Visage: Unveiling the Enigmatic Link between Phonemes and Facial Features Liao Qu*, Xianwei Zou*, Xiang Li*, Wendong Yan, Rita Singh, Bhiksha Raj InterSpeech, 2023 [paper] [code]

Self-supervised Multi-Modal Video Forgery Attack Detection Chenhui Zhao, Xiang Li, Rabhi Younes WCNC, 2023 [paper] [code]

Panoramic Video Salient Object Detection with Ambisonic Audio Guidance Xiang Li, Haoyuan Cao, Shijie Zhao, Junlin Li, Li Zhang, Bhiksha Raj AAAI, 2023 [paper] [visualization]

Forgery Attack Detection in Surveillance Video Streams Using Wi-Fi Channel State Information Yong Huang, Xiang Li, Wei Wang, Tao Jiang, Qian Zhang IEEE Transactions on Wireless Communication [paper]

Hybrid Instance-aware Temporal Fusion for Online Video Instance Segmentation Xiang Li, Jinglu Wang, Xiao Li, Yan Lu AAAI, 2022 [paper]

Video Instance Segmentation by Instance Flow Assembly Xiang Li, Jinglu Wang, Xiao Li, Yan Lu IEEE Transactions on Multimedia [paper]

Towards Cross-Modal Forgery Detection and Localization on Live Surveillance Videos Yong Huang, Xiang Li, Wei Wang, Tao Jiang, Qian Zhang IEEE INFOCOM, 2021 [paper]

Predicting Spatial Visualization Problems’ Difficulty Level from Eye-Tracking Data Xiang Li, Rabih Younes, Diana Bairaktarova, Qi Guo Sensors, 2020 [paper]

ActivityGAN: Generative Adversarial Networks for Data Augmentation in Sensor-Based Human Activity Recognition Xiang Li, Jinqi Luo, Rabih Younes DLHAR workshop @ Ubicomp (Best Paper Award), 2020 [paper]

Toward Data Augmentation and Interpretation in Sensor-Based Fine-Grained Hand Activity Recognition Jinqi Luo, Xiang Li, Rabih Younes ML4HAR workshop @ IJCAI, 2020 [paper]

Academic Activities

• Reviewer
  

  Conference: ICML, ICLR, NeurIPS, CVPR, ICCV, ECCV, EMNLP, NAACL, ACL, AAAI, ACM MM
  

• Mentorship
  

  Kai Qiu: MS at CMU (ongoing).
  

Liao Qu: MS at CMU. Now MLE at Tiktok.


Xianwei Zou: MS at CMU. Now PhD at UC Santa Barbara (UCSB).


Chenhui Zhao: MS at Duke. Now PhD at Duke University.


Zhaorun Chen: MS at Purdue. Now PhD at University of Chicago.


Fan Yang: MS at CMU. Now PhD at Ohio State University (OSU).