Xiao Guo

Xiao Guo (郭晓)

I am an applied scientist in Amazon TSI, working on the document legitimacy verification and forgery detection. I recently defended my Ph.D. at the CVLab of Michigan State University, advised by Prof. Xiaoming Liu. Before that, I spent years as a research programmer at USC/ISI, advised by Prof. Iacopo Masi and Prof. Wael AbdAlmageed. I obtain my master and bachelor at the University of Southern California and Wuhan University of Technology, respectively.

I have been a main contributor to several major U.S. government-sponsored projects, including MediFor, ODIN, and RED. Also, I spent two wonderful summer as an intern at Amazon in 2023 and 2024, working with Dr. Yue Rex Wu and Dr. Hongcheng Wang, respectively.

My research interests include:

Multi-modal Learning and Vision-Language Model: M2F2-Det (CVPR25-oral), DDVQA-BLIP (ECCV24), and MM-Det (NeurIPS24).
Image Generative Model: DenseFace (ArXiv24).
Responsible AI, e.g., Deepfake/Anti-spoofing: HiFi-Net (CVPR23), HiFi-Net++ (IJCV24), and FAS-Wrapper (ECCV22-oral).

Email / CV / Scholar / Github

News:

2026-02: Two works are accepted by CVPR26.
2026-01: Give a online lecture talk to the Vision and Language class at UNC-Chapel Hill.
2025-11: Defended my Ph.D. dissertation, and joined Amazon TSI as an applied scientist 🎉
2025-10: Two works are accepted by WACV26.
2025-06: Will attend ICCV25 Doctor Consortium.
2025-06: One paper is accepted by ICCV25.
2025-03: Our M2F2-Det is selected as an oral presentation (0.7% rate in the total submissions).
2025-02: Our M2F2-Det is accepted by CVPR25.
2024-12: Our DenseFace is posted on ArXiv.
2024-10: Our HiFi-Net++ is accepted by IJCV25.
2024-09: Two works (MM-Det and LGPN) are accepted by NeurIPS24.
2024-09: Completed my internship at Amazon One!
2023-09: Completed my internship at Amazon Alexa!
2022-05: Our ECCV22 paper is selected as an oral presentation (2.3% rate in the total submissions).

Selected Publications:

M2F2-Det teaser

Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
Xiao Guo, Xiufeng Song, Yue Zhang, Xiaohong Liu, Xiaoming Liu
CVPR (Oral Presentation), 2025
project page / code

/ arXiv

Formulating a deepfake detection task with Large Language Model.

HiFi-Net teaser

Hierarchical FineGrained Image Forgery Detection and Localization
Xiao Guo, Xiaohong Liu, Zhiyuan Ren, Steven Grosz, Iacopo Masi, Xiaoming Liu
CVPR, 2023; Extended version in IJCV, 2025 (IF=11.6)
code

/ arXiv

An image forgery detection and localization method for both digital manipulation and image editing domains.

MDFAS teaser

Multi-domain Learning for Updating Face Anti-spoofing Models
Xiao Guo, Yaojie Liu, Anil Jain, Xiaoming Liu
ECCV (Oral Presentation), 2022
code GitHub stars

GitHub stars

/ arXiv

A new model for multi-domain face anti-spoofing, which addresses the forgetting issue when learning new domain data.

Relation extraction teaser

Discourse-level Relation Extraction via Graph Pooling
I-Hung Hsu, Xiao Guo, Prem Natarajan, Nanyun Peng
AAAI Workshop on Deep Learning on Graphs (Best Paper Award), 2022
arXiv / Workshop Page

Other Publications:


	Common Sense Reasoning for Deepfake Detection Yue Zhang, Ben Colman, Xiao Guo, Ali Shahriyari, Gaurav Bharaj ECCV, 2024 code / arXiv Fine-tuning BLIP for deepfake detection VQA.
	On Learning Multi-Modal Forgery Representation for Diffusion Generated Video Detection Xiufeng Song, Xiao Guo, others , Xiaoming Liu Guangtao Zhai, Xiaohong Liu NeurIPS, 2024 code / arXiv A forgery video detection method based the LLama-2.
	Language-guided Hierarchical Finegrained Image Forgery Detection and Localization Xiao Guo, Xiaohong Liu, Iacopo Masi, Xiaoming Liu IJCV, 2025 code / arXiv Image Forgery Detection and Localization; An extension work of HiFi-Net (CVPR23).
	Dense-Face: Personalized Face Generation Model via Dense Annotation Prediction Xiao Guo, Manh Tran, Jiaxin Cheng, Xiaoming Liu arXiv, 2024 project page / code / arXiv A personalized face generation T2I diffusion model via dense landmarks prediction.
	Sea-CLIP: Mining Semantic-Aware Representations for Few-Shot Anomaly Detection with CLIP Xiao Guo, Zhimin Chen, Carlos D. Castillo, Hongcheng Wang, Xiaoming Liu WACV, 2026 arXiv Semantic-aware representations for few-shot anomaly detection with CLIP.
	Human motion prediction via learning local structure representations and temporal dependencies Xiao Guo, Jongmoo Choi AAAI 2019 code / arXiv
	Tracing Hyperparameter Dependencies for Model Parsing via Learnable Graph Pooling Network Xiao Guo, Vishal Asnani, Sijia Liu, Xiaoming Liu NeurIPS, 2024 arXiv Introduce a learnable graph pooling network for the model parsing.

Academic Services:

I regularly review papers for the following conferences and journals:

Conferences: CVPR, ICCV, ECCV, AAAI, NeurIPS, ICLR, etc.
Journals: T-PAMI, IJCV, TIFS, etc.