影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
35.09/100
前 18.8%
全站排名 #12,111
发表论文6 篇
平均评分
年均产出3.0 篇/年
Zihao Zhu
研究方向
Backdoor attack and defense · Large language model · Graph neural network
34
AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models
ICLR 2026Poster
一作29
Reliable Poisoned Sample Detection against Backdoor Attacks Enhanced by Sharpness Aware Minimization
ICLR 2026Poster
三作24
To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models
ICLR 2026Rejected
一作19
VDC: Versatile Data Cleanser based on Visual-Linguistic Inconsistency by Multimodal Large Language Models
ICLR 2024Poster
一作27
Boosting Backdoor Attack with A Learnable Poisoning Sample Selection Strategy
ICLR 2024Rejected
一作23
Robust Backdoor Attack with Visible, Semantic, Sample-specific and Compatible Triggers
ICLR 2024Rejected
三作