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
35.4/100
前 18.5%
全站排名 #11,902
发表论文5 篇
平均评分
年均产出2.5 篇/年
Samet Demir
研究方向
theory of deep learning · random matrix theory · high-dimensional asymptotics · random features · Gaussian equivalence · universality · learning dynamics · mean-field analysis
23
In-Context Learning under Distribution Shift: Optimal Attention Temperature for Transformers
NeurIPS 2025Rejected
一作22
How Data Mixing Shapes In-Context Learning: Asymptotic Equivalence for Transformers with MLPs
NeurIPS 2025Poster
一作22
Asymptotic Analysis of Two-Layer Neural Networks after One Gradient Step under Gaussian Mixtures Data with Structure
ICLR 2025Poster
一作37
Random Features Outperform Linear Models: Effect of Strong Input-Label Correlation in Spiked Covariance Data
ICLR 2025Rejected
一作合作者 (1)