α
Research
Alpha Leak
Conferences
Topics
Top Authors
Rankings
Browse All
EN
中
Home
/
Authors
/
Barbara Plank
Barbara Plank
4
papers
45
total citations
papers (4)
Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation
ICLR 2025
arXiv
26
citations
Position: Insights from Survey Methodology can Improve Training Data
ICML 2024
arXiv
11
citations
Refusal Direction is Universal Across Safety-Aligned Languages
NEURIPS 2025
arXiv
5
citations
Mind the Uncertainty in Human Disagreement: Evaluating Discrepancies Between Model Predictions and Human Responses in VQA
AAAI 2025
arXiv
3
citations