Dana Arad
Scholar

Dana Arad

Google Scholar ID: A363uwwAAAAJ
PhD Student, Technion
NLPInterpretabilityVision-Language
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
61
 
H-index
4
 
i10-index
2
 
Publications
8
 
Co-authors
0
 
Contact
Emaildanaarad@campus.technion.ac.ilTwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
8 items
Mechanisms of Prompt-Induced Hallucination in Vision-Language Models
arXiv.org · 2026
Cited
0
Findings of the BlackboxNLP 2025 Shared Task: Localizing Circuits and Causal Variables in Language Models
2025
Cited
0
BlackboxNLP-2025 MIB Shared Task: Improving Circuit Faithfulness via Better Edge Selection
2025
Cited
0
HACK: Hallucinations Along Certainty and Knowledge Axes
2025
Cited
0
CRISP: Persistent Concept Unlearning via Sparse Autoencoders
2025
Cited
0
Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs
2025
Cited
0
SAEs Are Good for Steering -- If You Select the Right Features
2025
Cited
0
MIB: A Mechanistic Interpretability Benchmark
2025
Cited
0
Resume (English only)
Co-authors
0 total
Co-authors: 0 (list not available)