AI · Multimedia · Distributed Learning

About me

I am a Ph.D. student at The University of Tokyo, working with Prof. Hideya Ochiai in the Esaki & Ochiai Lab.

My research focuses on multimodal federated learning, with an emphasis on missing modalities and heterogeneous clients. I also work on multimodal understanding and LLM-driven smart contract security.

Previously, I received my M.Eng. from Beijing Institute of Technology, advised by Prof. Keke Gai, and my B.Eng. and B.Econ. from Jilin University. I was a research assistant with Prof. Zitong Yu at Great Bay University and a visiting student with Prof. Giulia Boato at the University of Trento.

Multimodal LearningFederated LearningLarge Language ModelsTrustworthy AI

Research highlights

FedTaste accepted to ACM Multimedia 2026.

MViR at ICASSP 2026: multi-view representations for fake news detection.

LLAMA in IEEE Transactions on Information Forensics and Security: multi-feedback smart contract fuzzing with LLM-guided seed generation.

LENS accepted to ICPADS 2026: end-to-end LLM-enhanced smart contract fuzzing for blockchain security.

Visiting student at the University of Trento, working with Prof. Giulia Boato.

Research visit to Great Bay University.

Started my Ph.D. at The University of Tokyo.

Publications

Google Scholar ↗
ACM MM2026

FedTaste: Topology-Aware Structural Transfer for Multimodal Federated Learning with Missing Modalities

H. Liang, J. Zhang, H. Ochiai

ACM MM 2026Accepted

Topology-aware structural transfer for collaborative learning with missing modalities.

Paper ↗
ICASSP2026

MViR: Multi-View Visual-Semantic Representation for Fake News Detection

H. Liang, X. Su, J. Wang, C. Chen, Z. Yu

ICASSP 2026

Multi-view visual-semantic representations for multimodal fake news detection.

Paper ↗Code ↗
IEEE TIFS

LLAMA: Multi-Feedback Smart Contract Fuzzing Framework with LLM-Guided Seed Generation

K. Gai, H. Liang, J. Yu, L. Zhu, D. Niyato

IEEE TIFS

Combining LLM-guided seed generation with multi-feedback optimization for smart contract fuzzing.

Paper ↗
ICPADS2026

LENS: End-to-End LLM-Enhanced Smart Contract Fuzzing for Blockchain Security

H. Liang, J. Chen, et al.

ICPADS 2026Accepted

GCCE2026

Dual-Level Prototype Alignment for Heterogeneous Multimodal Federated Learning

H. Liang, H. Ochiai

GCCE 2026Accepted

GCCE2026

Blockchain-Based Trustworthy Multimodal Federated Learning

H. Liang, H. Ochiai

GCCE 2026Accepted

IEEE IoT-J2026

FELT: Federated Ensemble Learning for Long-Tailed IoT Data via Communication-Efficient Private Voting

C. Chen, S. Ye, H. Liang, et al.

IEEE IoT-J 2026

DOI ↗
ACL2026

MAVIS: Multi-Agent Video Retrieval via Structured Video Understanding

J. Zhang, Q. Ye, H. Zhou, H. Liang, et al.

ACL 2026

arXiv ↗
ACM MM2026

AffectAgent: Collaborative Multi-Agent Reasoning for Retrieval-Augmented Multimodal Emotion Recognition

Z. Wang, Z. Yu, Y. Zhu, B. Zhao, H. Liang, et al.

ACM MM 2026Accepted

arXiv ↗
ACM MM2025

Dynamic Analysis and Adaptive Discriminator for Fake News Detection

X. Su, Z. Yu, Y. Cui, A. Liu, X. Lin, Y. Wang, H. Liang, et al.

ACM MM 2025

Paper ↗
KSEM2024

KDTSS: A Blockchain-Based Scheme for Knowledge Data Traceability and Secure Sharing

H. Liang, Y. Guo, et al.

KSEM 2024

KSEM2024

DICES: Diffusion-Based Contrastive Learning with Knowledge Graphs for Recommendation

H. Dong†, H. Liang†, et al.

KSEM 2024

Blockchains2024

Zk-SNARKs-Based Anonymous Payment Channel in Blockchain

Y. Guo, H. Liang*, et al.

Blockchains 2024

SmartCloud2023

A Blockchain-Based Hierarchical Storage Method for Supply Chain Data

H. Liang, Y. Guo, K. Gai

SmartCloud 2023

† Equal contribution · * Corresponding author

Manuscripts under review (5)

Submission venues are listed for context; these manuscripts are not accepted publications.

Learning Cross-Modal Generative Priors for Multimodal Federated Learning with Missing Modalities

Haochen Liang, Chaomeng Chen, Hui Ma, Yawen Cui, Jie Zhang, Zeheng Wang, Junzhe Cao, Jing Yang, Zitong Yu

ICLR 2027 · Under review

Empowering Large Vision-Language Models to Bridge Open-Set Domain Gaps in Video Retrieval

J. Zhang, H. Liang, et al.

AAAI 2027 · Under review

Semantic Manifold Completion for Partially Relevant Video Retrieval

J. Zhang, Q. Ye, H. Liang, et al.

AAAI 2027 · Under review

FedXRef: Learning Cross-Modal Prototype References for Heterogeneous Multimodal Federated Learning

Haochen Liang, Chaomeng Chen, Jie Zhang, Hideya Ochiai*

ICASSP 2027 · Under review

Redistributing the Benefits of Collaborative Critical-Care AI Across Data-Unequal Hospital Sites

C. Chen, Q. Li, J. Gu, X. Lin, Y. Wang, J. Zhu, K. Wang, H. Liang, et al.

Nature Communications · Under review

Education & experience

The University of Tokyo

Ph.D. in Information and Communication Engineering · Esaki & Ochiai Lab
Advisor: Prof. Hideya Ochiai

University of Trento

Visiting student · Advisor: Prof. Giulia Boato

Great Bay University

Research assistant · Advisor: Prof. Zitong Yu

Beijing Institute of Technology

M.Eng. in Electronic Information · Advisor: Prof. Keke Gai

Jilin University

B.Eng. in Mechanical Engineering & B.Econ. (dual degree)

Academic service

Journal reviewer

IEEE Transactions on Information Forensics and Security · Pattern Recognition

Conference reviewer

ICASSP 2025 · IJCB 2025 · KSEM 2024 · BigDataSecurity 2024 · SmartCloud 2023 · SmartCom 2023

Honors & awards

  • IEEE Computer Society WEIGD Student Support Fund, 2026
  • UTokyo Self-Directed and Integrated Project Research Funding, 2025–2026
  • Outstanding Graduate, Beijing Institute of Technology, 2025
  • Student Innovation Funding Program, 2024–2025
  • Graduate Academic Scholarship, Beijing Institute of Technology, 2022–2024
  • First Prize, Chinese Mathematics Competition for College Students, 2022