Author

Tianyu Du

Observed name: Du, Tianyu

Institution: Zhejiang University

Observed institution: Zhejiang University

OpenAlex ID: A5139755055

LSHFed: Robust and Communication-Efficient Federated Learning with Locally-Sensitive Hashing Gradient Mapping

2026 · Guanjie Cheng, Mengzhen Yang, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2

PDF

DP-GenG: Differentially Private Dataset Distillation Guided by DP-Generated Data

2026 · Shuo Shi, Jinghuai Zhang, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 1

PDF

FedAU2: Attribute Unlearning for User-Level Federated Recommender Systems with Adaptive and Robust Adversarial Training

2026 · Yuyuan Li, Junjie Fang, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 1

PDF

Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors

2026 · Rui Yin, Tianxu Han, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting · 1

PDF

StateWise: Diagnosing and Repairing Persistent Operational State Before Agent Actions

2026 · Peng, Yongyuan, Feng, Zhou, et al. · arXiv (Cornell University)

PDF

Safety in Self-Evolving Agents: A Survey

2026 · Jiahao Chen, Zhou Feng, et al. · arXiv (Cornell University)

PDF

The Boy Who Cried Wolf: Adversarial Misclassification of Safe Inputs as Unsafe in Multimodal Guardrails

2026 · Shi, Shuo, Yin, Rui, et al. · arXiv (Cornell University)

PDF

Flipping Minds by Flipping Bits: Decision-Level Cognitive Hijacking of Large Language Models

2026 · Yan, Yu, Chen, Jiahao, et al. · arXiv (Cornell University)

PDF

Lilith: Backdoor Generalization under Training-Inference Trigger Shift

2026 · Feng, Zhou, Chen, Jiahao, et al. · arXiv (Cornell University)

PDF

Understanding and Mitigating Prompt Leaking Attacks in Real-World LLM-Based Applications

2026 · Yang, Yong, Fu, Chong, et al. · arXiv (Cornell University)

PDF

Customization under Fire: Plugin Poisoning in Text-to-Image Ecosystem

2026 · Chen, Jiahao, He, Xing, et al. · arXiv (Cornell University)

PDF

ACIArena: Toward Unified Evaluation for Agent Cascading Injection

2026 · Association for Computational Linguistics 2026, An, Hengyu, et al. · Underline Science Inc.

“I See What You Did There”: Can Large Vision-Language Models Understand Multimodal Puns?

2026 · Association for Computational Linguistics 2026, Du, Tianyu, et al. · Underline Science Inc.

Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors

2026 · Association for Computational Linguistics 2026, Du, Tianyu, et al. · Underline Science Inc.

Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors

2026 · Yin, Rui, Han, Tianxu, et al. · arXiv (Cornell University)

PDF

ACIArena: Toward Unified Evaluation for Agent Cascading Injection

2026 · An, Hengyu, Li, Minxi, et al. · arXiv (Cornell University)

PDF

"I See What You Did There": Can Large Vision-Language Models Understand Multimodal Puns?

2026 · Xu, Naen, Sheng, Jiayi, et al. · arXiv (Cornell University)

PDF

HogVul: Black-box Adversarial Code Generation Framework Against LM-based Vulnerability Detectors

2026 · Jingxiao Yang, Ping He, et al. · Proceedings of the AAAI Conference on Artificial Intelligence

PDF

From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI Agents

2026 · Zhang, Xiaolei, Zhou, Lu, et al. · arXiv (Cornell University)

PDF

When Agents "Misremember" Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems

2026 · Xu, Naen, An, Hengyu, et al. · arXiv (Cornell University)

PDF

FraudShield: Knowledge Graph Empowered Defense for LLMs against Fraud Attacks

2026 · Xu, Naen, Zhang, Jinghuai, et al. · arXiv (Cornell University)

PDF

HogVul: Black-box Adversarial Code Generation Framework Against LM-based Vulnerability Detectors

2026 · Yang, Jingxiao, He, Ping, et al. · arXiv (Cornell University)

PDF

FedAU2: Attribute Unlearning for User-Level Federated Recommender Systems with Adaptive and Robust Adversarial Training

2026 · Association for Artificial Intelligence 2026, Chen, Chaochao, et al. · Underline Science Inc.

Bridging the Copyright Gap: Do Large Vision-Language Models Recognize and Respect Copyrighted Content?

2026 · Association for Artificial Intelligence 2026, An, Hengyu, et al. · Underline Science Inc.

DP-GenG: Differentially Private Dataset Distillation Guided by DP-Generated Data

2026 · Association for Artificial Intelligence 2026, Du, Tianyu, et al. · Underline Science Inc.

Works count: 34

Citation count: 5

H-index: 1

I10-index: