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
39.47/100
前 14.8%
全站排名 #9,526
发表论文4 篇
平均评分
年均产出2.0 篇/年
Yiwei Chen
研究方向
Trustworthy ML · Vision Language Model · Multi-modal Learning · Few-shot Learning
22
Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearning
ICLR 2026Poster
一作27
Unlearning Isn't Invisible: Detecting Unlearning Traces in LLMs from Model Outputs
ICLR 2026Poster
一作19
Leak@$k$: Unlearning Does Not Make LLMs Forget Under Probabilistic Decoding
ICLR 2026Rejected
三作