Hemanth Saratchandran
~Hemanth_Saratchandran1
8
论文总数
4.0
年均投稿
平均评分
接收情况4/8
会议分布
ICLR
6
NeurIPS
2
发表论文 (8 篇)
20256 篇
4
Rethinking Softmax: Self-Attention with Polynomial Activations
ICLR 2025withdrawn
4
Spectral Conditioning of Attention Improves Transformer Performance
NeurIPS 2025Poster
5
Sampling Theory and Overparameterization: Shaping Loss Landscapes in $\ell^2$ Regression
ICLR 2025withdrawn
4
Efficient Learning with Sine-Activated Low-Rank Matrices
ICLR 2025Poster
4
Structured Initialization for Vision Transformers
NeurIPS 2025Poster
4
RandLoRA: Full rank parameter-efficient fine-tuning of large models
ICLR 2025Poster