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
50.74/100
前 8.2%
全站排名 #5,292
发表论文6 篇
平均评分
年均产出2.0 篇/年
Aryaman Arora
研究方向
computational linguistics · mechanistic interpretability · South Asian languages · language modelling · neural network architectures
11
AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
ICML 2025Spotlight
二作17
Improved Representation Steering for Language Models
NeurIPS 2025Spotlight
三作14
Bayesian scaling laws for in-context learning
COLM 2025Poster
一作22
Bayesian scaling laws for in-context learning
ICLR 2025Rejected
一作