影响力指数

41.03/100

前 13.7%

全站排名 #8,838

发表论文5 篇

平均评分5.4

年均产出1.7 篇/年

Elan Rosenfeld

Researcher@Google Research·美国·OpenReview

Hidden Breakthroughs in Language Model Training

ICLR 2026Poster

Transformers tend to memorize geometrically; it is unclear why.

ICLR 2026Withdrawn

Loss in the Crowd: Hidden Breakthroughs in Language Model Training

ICLR 2025Rejected

合作者 (10)

Sara Kangaslahti

Pradeep Kumar Ravikumar

Sorawit Saengkyongam

Andrej Risteski

博士导师1 篇