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
33.45/100
前 20.6%
全站排名 #13,283
发表论文4 篇
平均评分
年均产出2.0 篇/年
Tin Sum Cheng
研究方向
kernel learning · neural networks · discrete optmization · graph theory · integer programming
20
Optimizer Choice Matters For The Emergence of Neural Collapse
ICLR 2026Poster
二作21
Spectral Analysis of Molecular Kernels: When Richer Features Do Not Guarantee Better Generalization
ICLR 2026Rejected
二作4
Unpacking Softmax: How Logits Norm Drives Representation Collapse, Compression and Generalization
ICLR 2026Withdrawn
三作