Omid Saremi
~Omid_Saremi1
7
论文总数
7.0
年均投稿
平均评分
接收情况6/7
会议分布
ICLR
5
NeurIPS
2
发表论文 (7 篇)
20247 篇
5
When can transformers reason with abstract symbols?
ICLR 2024Poster
4
Adaptivity and Modularity for Efficient Generalization Over Task Complexity
ICLR 2024Rejected
5
How JEPA Avoids Noisy Features: The Implicit Bias of Deep Linear Self Distillation Networks
NeurIPS 2024Poster
4
Vanishing Gradients in Reinforcement Finetuning of Language Models
ICLR 2024Poster
3
LiDAR: Sensing Linear Probing Performance in Joint Embedding SSL Architectures
ICLR 2024Spotlight
4
What Algorithms can Transformers Learn? A Study in Length Generalization
ICLR 2024Poster
4
How Far Can Transformers Reason? The Globality Barrier and Inductive Scratchpad
NeurIPS 2024Poster