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
61.63/100
前 4.4%
全站排名 #2,838
发表论文8 篇
平均评分
年均产出2.7 篇/年
Barbara Plank
研究方向
human-centric NLP · robustness and trustworthiness · uncertainty · human label variation · disagreement · multilingual · low-resource NLP
20
Is it Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort
ICLR 2026Oral
三作17
Compositional-ARC: Assessing Systematic Generalization in Abstract Spatial Reasoning
ICLR 2026Poster
通讯14
ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior
ICLR 2026Rejected
22
Refusal Direction is Universal Across Safety-Aligned Languages
NeurIPS 2025Poster
通讯23
Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation
ICLR 2025Poster
通讯-1
Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models
ICLR 2025Desk Rejected
三作