Zhiyuan Chang
Scholar

Zhiyuan Chang

Google Scholar ID: DWk0W58AAAAJ
Institute of Software Chinese Academy of Science
LLM SecurityMultimodal TestingRequirements Engineering
Google Scholar↗
Citations & Impact
All-time
Citations
88
 
H-index
4
 
i10-index
1
 
Publications
12
 
Co-authors
6
list available
Contact
No contact links provided.
Publications
8 items
Emerging from Ground: Addressing Intent Deviation in Tool-Using Agents via Deriving Real Calls into Virtual Trajectories
2026
Cited
0
Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning
arXiv.org · 2026
Cited
0
All Changes May Have Invariant Principles: Improving Ever-Shifting Harmful Meme Detection via Design Concept Reproduction
arXiv.org · 2026
Cited
0
Learning from Mistakes: Enhancing Harmful Meme Detection via Misjudgment Risk Patterns
2025
Cited
0
Butterfly Effects in Toolchains: A Comprehensive Analysis of Failed Parameter Filling in LLM Tool-Agent Systems
2025
Cited
0
One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems
2025
Cited
0
Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System
2025
Cited
0
What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context
arXiv.org · 2024
Cited
0
Resume (English only)
Co-authors
6 total
Junjie Wang
Junjie Wang
Institute of Software, Chinese Academy of Sciences
Qing Wang
Qing Wang
Institute of Software Chinese Academy of Sciences
Yang Liu
Yang Liu
Nanyang Technological University
Yi Liu
Yi Liu
AI Research @ Quantstamp | PhD @ NTU | BEng @ SUSTech
Shoubin Li
Shoubin Li
Institute of Software, Chinese Academy of Sciences
Xiaojun Jia
Xiaojun Jia
Nanyang Technological University