Victor Shea-Jay Huang's Homepage

Short Bio

Hi 🤗, I’m Victor Shea-Jay Huang (Chinese: 黄希杰(Xijie Huang); French: Victor Ché-Jay Huang).

I am an undergraduate student at Beihang University, expected to graduate in 2026.

I am currently a research intern at ByteDance TikTok. Previously, I worked as a research intern at Shanghai AI Laboratory and Baichuan AI.

If you are interested in joining the project or exploring collaboration opportunities, please reach out to me via email at jeix782@gmail.com.

Research Interest

I have a strong interest in all topics related to AI integrated with natural scientific theories and am actively exploring and expanding the boundaries of current AI technology. My efforts are primarily focused on 3 directions:

Multimodal: Centered on generative model, multimodal learning, and efficient multimodal technologies for enhanced performance and integration.
LLM: Focused on core LLM advancements, including reasoning, scaling laws, and foundational research to push technological boundaries.
Trustworthy AI: Identify and address critical challenges in trustworthy AI, focusing on improving interpretability, safety, privacy, etc.

News

[Jun. 2025] 🎉🎉 Our JPS have been accepted to ACM MM 2025!(CCF-A)
[Jun. 2025] 🎉🎉 Our Vision-to-Music Generation have been accepted to ISMIR 2025!(Top conferences in music AI)
[Apr. 2025] 🎉🎉 Lumina-mGPT-2.0 github repo released .
[Mar. 2025] 🎉🎉 Awesome Vision-to-Music github repo released .
[Feb. 2025] 🎉🎉 I've been accepted into the SMC Student Scholarship!
[Feb. 2025] 🎉🎉 Our 2 papers Beyond Sight and Scaling Laws for Video DiT have been accepted to CVPR 2025!(CCF-A, Acceptance ratio 22.1%)
[Jan. 2025] 🎉🎉 AISafetyLab github repo released .
[Dec. 2024] 🎉🎉 I've been accepted into the AAAI-25 Student Scholarship!
[Dec. 2024] 🎉🎉 Our 2 papers Medical MLLM is Vulnerable and AGFSync have been accepted to AAAI 2025!(CCF-A, Acceptance ratio 23.4%)
[Nov. 2024] 🎉🎉 Welcome to my YouTube channel @VSJH2003!
[Aug. 2024] 🎉🎉 I have founded MantaAI: Modified Approaches in Natural Sciences to Advance Artificial Intelligence. Researchers from around the world who share the same vision are welcome to join.

Education

B.Eng., 2022 - 2026, In Institute of Automation Science and Electrical Engineering, Beihang University, Beijing, China
B.Sci., 2021 - 2022, In l’École Centrale de Pékin, Beijing, China

Industry Experience

July 2025 - Present, TikTok

Research intern. Beijing, China

Jan 2025 - May 2025, OpenGVLab

Research intern. Shanghai, China

Sep 2024 - Dec 2024

Research intern. Beijing, China

Invited Talks

2025.1, Beijing, China. AI Time Youth PhD Talk-AAAI 2025

Selected Awards

2025.2, SMC Scholarship
2024.12, AAAI 2025 Student Scholarship

Services

Conference Reviewers

Publications [Google Scholar]

AAAI

Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models

Victor Shea-Jay Huang, Xinyuan Wang, Hantao Zhang, Yinghao Zhu, Jiawen Xi, Jingkun An, Hao Wang, Hao Liang, Chengwei Pan

AAAI 2025

PDF Code Arxiv BibTex Published
ICML

BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models

Victor Shea-Jay Huang*, Xinyuan Wang*, Renmiao Chen, Hao Wang, Chengwei Pan, Lei Sha, Minlie Huang

ICML 2025 NewInML Workshop

PDF Code Arxiv BibTex Published
CVPR

Beyond Sight: Towards Cognitive Alignment in LVLM via Enriched Visual Knowledge

Yaqi Zhao, Yuanyang Yin, Lin Li, Mingan Lin, Victor Shea-Jay Huang, Siwei Chen, Weipeng Chen, Baoqun Yin, Zenan Zhou, Wentao Zhang

CVPR 2025

PDF Arxiv BibTex Published.
CVPR

Towards Precise Scaling Laws for Video Diffusion Transformers

Yuanyang Yin, Yaqi Zhao, Mingwu Zheng, Ke Lin, Jiarong Ou, Rui Chen, Victor Shea-Jay Huang, Jiahao Wang, Xin Tao, Pengfei Wan, Di ZHANG, Baoqun Yin, Wentao Zhang, Kun Gai

CVPR 2025

PDF Arxiv BibTex Published.
AAAI

AGFSync: Leveraging AI Feedback for Preference Optimization in Text-to-Image Generation

Jingkun An, Yinghao Zhu, Zongjian Li, Haoran Feng, Victor Shea-Jay Huang, Bohua Chen,Yemin Shi, Chengwei Pan

AAAI 2025

PDF Arxiv BibTex Published
ISMIR

Vision-to-Music Generation: A Survey

Zhaokai Wang, Chenxi Bao, Le Zhuo, Jingrui Han, Yang Yue, Yihong Tang, Victor Shea-Jay Huang, Yue Liao

ISMIR 2025

PDF Code Arxiv Published
AAAI

Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models

Victor Shea-Jay Huang, Xinyuan Wang, Hantao Zhang, Yinghao Zhu, Jiawen Xi, Jingkun An, Hao Wang, Hao Liang, Chengwei Pan

AAAI 2025

PDF Code Arxiv BibTex Published
CVPR

Beyond Sight: Towards Cognitive Alignment in LVLM via Enriched Visual Knowledge

Yaqi Zhao, Yuanyang Yin, Lin Li, Mingan Lin, Victor Shea-Jay Huang, Siwei Chen, Weipeng Chen, Baoqun Yin, Zenan Zhou, Wentao Zhang

CVPR 2025

PDF Arxiv BibTex Published.
CVPR

Towards Precise Scaling Laws for Video Diffusion Transformers

Yuanyang Yin, Yaqi Zhao, Mingwu Zheng, Ke Lin, Jiarong Ou, Rui Chen, Victor Shea-Jay Huang, Jiahao Wang, Xin Tao, Pengfei Wan, Di ZHANG, Baoqun Yin, Wentao Zhang, Kun Gai

CVPR 2025

PDF Arxiv BibTex Published.
AAAI

AGFSync: Leveraging AI Feedback for Preference Optimization in Text-to-Image Generation

Jingkun An, Yinghao Zhu, Zongjian Li, Haoran Feng, Victor Shea-Jay Huang, Bohua Chen,Yemin Shi, Chengwei Pan

AAAI 2025

PDF Arxiv BibTex Published
Arxiv

Document Parsing Unveiled: Techniques, Challenges, and Prospects for Structured Information Extraction

Qintong Zhang*, Victor Shea-Jay Huang*, Bin Wang, Junyuan Zhang, Zhengren Wang, Hao Liang, Shawn Wang, Matthieu Lin, Conghui He, Wentao Zhang

Arxiv Preprint

PDF Arxiv BibTex Preprint
Arxiv

Vision-to-Music Generation: A Survey

Zhaokai Wang, Chenxi Bao, Le Zhuo, Jingrui Han, Yang Yue, Yihong Tang, Victor Shea-Jay Huang, Yue Liao

Arxiv Preprint

PDF Code Arxiv Under review.
Arxiv

Trans4D: Realistic Geometry-Aware Transition for Compositional Text-to-4D Synthesis

Bohan Zeng, Ling Yang, Siyu Li, Jiaming Liu, Zixiang Zhang, Victor Shea-Jay Huang, Juanxi Tian, Kaixin Zhu, yongzhen.gyz, Fu-Yun Wang, Minkai Xu, Stefano Ermon, Wentao Zhang

Arxiv Preprint

PDF Arxiv BibTex Under review.
Arxiv

SynthVLM: High-Efficiency and High-Quality Synthetic Data for Vision Language Models

Zheng Liu, Hao Liang, Victor Shea-Jay Huang, Wentao Xiong, Qinhan Yu, Linzhuang Sun, Chong Chen, Conghui He, Bin Cui, Wentao Zhang

Arxiv Preprint

PDF Code Arxiv BibTex Under review.
Arxiv

KeyVideoLLM: Towards Large-scale Video Keyframe Selection

Zheng Liu, Hao Liang, Victor Shea-Jay Huang, Wentao Xiong, Qinhan Yu, Linzhuang Sun, Chong Chen, Conghui He, Bin Cui, Wentao Zhang

Arxiv Preprint

PDF Arxiv BibTex Under review.
Arxiv

AISafetyLab: A Comprehensive Framework for AI Safety Evaluation and Improvement

Zhexin Zhang*, Leqi Lei*, Junxiao Yang*, Victor Shea-Jay Huang*, Yida Lu, Shiyao Cui, Renmiao Chen, Qinglin Zhang, Xinyuan Wang, Hao Wang, Hao Li, Xianqi Lei, Chengwei Pan, Lei Sha, Hongning Wang, Minlie Huang

Arxiv Preprint

PDF Code Arxiv Preprint
Arxiv

How Should We Enhance the Safety of Large Reasoning Models: An Empirical Study

Zhexin Zhang*, Xian Qi Loye*, Victor Shea-Jay Huang, Junxiao Yang, Qi Zhu, Shiyao Cui, Fei Mi, Lifeng Shang, Yingkang Wang, Hongning Wang, Minlie Huang

Arxiv Preprint

PDF Code Arxiv Preprint
ICML

BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models

Victor Shea-Jay Huang*, Xinyuan Wang*, Renmiao Chen, Hao Wang, Chengwei Pan, Lei Sha, Minlie Huang

ICML 2025 Workshop

PDF Code Arxiv BibTex Published
Arxiv

Synth-Empathy: Towards High-Quality Synthetic Empathy Data

Hao Liang, Linzhuang Sun, Jingxuan Wei, Victor Shea-Jay Huang, Linkun Sun, Bihui Yu, Conghui He, Wentao Zhang

Arxiv Preprint

PDF Code Arxiv BibTex Under review.
AAAI

Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models

Victor Shea-Jay Huang, Xinyuan Wang, Hantao Zhang, Yinghao Zhu, Jiawen Xi, Jingkun An, Hao Wang, Hao Liang, Chengwei Pan

AAAI 2025

PDF Code Arxiv BibTex Published
Arxiv

AISafetyLab: A Comprehensive Framework for AI Safety Evaluation and Improvement

Zhexin Zhang*, Leqi Lei*, Junxiao Yang*, Victor Shea-Jay Huang*, Yida Lu, Shiyao Cui, Renmiao Chen, Qinglin Zhang, Xinyuan Wang, Hao Wang, Hao Li, Xianqi Lei, Chengwei Pan, Lei Sha, Hongning Wang, Minlie Huang

Arxiv Preprint

PDF Code Arxiv Preprint
Arxiv

How Should We Enhance the Safety of Large Reasoning Models: An Empirical Study

Zhexin Zhang*, Xian Qi Loye*, Victor Shea-Jay Huang, Junxiao Yang, Qi Zhu, Shiyao Cui, Fei Mi, Lifeng Shang, Yingkang Wang, Hongning Wang, Minlie Huang

Arxiv Preprint

PDF Code Arxiv Preprint
Arxiv

BlackDAN: A Black-Box Multi-Objective Approach for Effective and Contextual Jailbreaking of Large Language Models

Xinyuan Wang*, Victor Shea-Jay Huang*, Renmiao Chen, Hao Wang, Chengwei Pan, Lei Sha, Minlie Huang

Arxiv Preprint

PDF Code Arxiv BibTex Under review.