Get started
Author

Ziyi Zhou

Attack the Messages, Not the Agents: A Multi-round Adaptive Stealthy Tampering Framework for LLM-MAS
2026 · Bingyu Yan, Xiaoming Zhang, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2
On the Feasibility of Using MultiModal LLMs to Execute AR Social Engineering Attacks
2026 · Ting Bi, Chenghang Ye, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2
PDF
Reasoning through Evolution: Automatic Meta-path Discovery for LLM-based Fake News Detection
2026 · Zhou, Ziyi, Zhang, Xiaoming, et al. · arXiv (Cornell University)
PDF
Benign Alone, Harmful Together: Exploiting Experience Composition in Self-Evolving LLM Agents
2026 · Yan, Bingyu, Zhang, Xiaoming, et al. · arXiv (Cornell University)
PDF
Evo-Attacker: Memory-Augmented Reinforcement Learning for Long-Horizon Tool Attacks on LLM-MAS
2026 · Association for Computational Linguistics 2026, Hei, Yiming, et al. · Underline Science Inc.
Beyond Static Persona Consistency: Dynamic Persona Coherence in LLM Role-Playing
2026 · Association for Computational Linguistics 2026, Guan, Zhenyu, et al. · Underline Science Inc.
Evo-Attacker: Memory-Augmented Reinforcement Learning for Long-Horizon Tool Attacks on LLM-MAS
2026 · Yan, Bingyu, Zhang, Xiaoming, et al. · arXiv (Cornell University)
PDF
PhySE: A Psychological Framework for Real-Time AR-LLM Social Engineering Attacks
2026 · Yu, Tianlong, Yang, Yang, et al. · arXiv (Cornell University)
PDF
GeWu: A Culturally-Grounded Chinese Benchmark for Multi-Stage Social Bias Evaluation in Large Language Models
2026 · Yi Lin, Ziyi Zhou, et al. · Proceedings of the AAAI Conference on Artificial Intelligence
PDF
On the Feasibility of Using MultiModal LLMs to Execute AR Social Engineering Attacks
2026 · Association for Artificial Intelligence 2026, Bi, Ting, et al. · Underline Science Inc.
GeWu: A Culturally-Grounded Chinese Benchmark for Multi-Stage Social Bias Evaluation in Large Language Models
2026 · Association for Artificial Intelligence 2026, Gao, Jiashi, et al. · Underline Science Inc.
Evo-Attacker: Memory-Augmented Reinforcement Learning for Long-Horizon Tool Attacks on LLM-MAS
2026 · Bingyu Yan, Xiaoming Zhang, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting
PDF
F-PRIMO: Publicly Verifiable Computation over Multi-Owner Filtered Outsourced Data
2026 · Min Xie, Qingyuan Jia, et al. · IEEE Transactions on Dependable and Secure Computing
Works count: 16
Citation count: 4
H-index: 2
I10-index:
Year
Topic
Type