TY  - RPRT
TI  - Position: Zeroth-Order Optimization in Deep Learning Is Underexplored, Not Underpowered
AU  - Sijia Liu
AU  - Yicheng Lang
AU  - Soumyadeep Pal
AU  - Changsheng Wang
AU  - Yancheng Huang
AU  - Chongyu Fan
AU  - James Diffenderfer
AU  - Bhavya Kailkhura
AU  - Yihua Zhang
PY  - 2026
UR  - https://arxiv.org/abs/2605.15622
ID  - 2605.15622
ER  -