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
38.68/100
前 15.4%
全站排名 #9,905
发表论文6 篇
平均评分
年均产出3.0 篇/年
Narine Kokhlikyan
研究方向
Model Interpretability · Machine Learning · Memorization
13
CIMemories: A Compositional Benchmark For Contextual Integrity In LLMs
ICLR 2026Poster
三作12
Hallucination Reduction with CASAL: Contrastive Activation Steering for Amortized Learning
ICLR 2026Poster
13
Z0-Inf: Zeroth Order Approximation for Data Influence
ICLR 2026Rejected
一作21
How much can language models memorize?
ICLR 2026Rejected
三作14
RL Is a Hammer and LLMs Are Nails: A Simple Reinforcement Learning Recipe for Strong Prompt Injection
ICLR 2026Rejected