Get started
Author

Amit LeVi

Observed names: LeVi, Amit, Levi, Amit
OpenAlex ID: A5124977942
Activation Steering for Masked Diffusion Language Models
2025 · Shnaidman, Adi, Feiglin, Erin, et al. · arXiv (Cornell University) · 12
PDF
Mirage Probes: How Vision Models Fake Visual Understanding
2026 · Ben-Levi, Daniel, Goldfeder, Judah, et al. · arXiv (Cornell University) · 6
PDF
Internal-State Probes Read the Situation, Not the Action: Three Negative Results for Pre-Action Misalignment Monitoring
2026 · Fomin, Max, David, Elad, et al. · arXiv (Cornell University) · 4
PDF
Extracting Recurring Vulnerabilities from Black-Box LLM-Generated Software
2026 · Kordonsky, Tomer, LeVi, Amit, et al. · arXiv (Cornell University) · 1
PDF
Generalization Is Stability, Not Accuracy: Multi-Axis Evaluation of LLMs
2026 · Omar, Nagham, Jabarin, Mahmoud, et al. · arXiv (Cornell University)
PDF
Unsupervised Features Mining via Activation Geometry
2026 · LeVi, Amit, David, Elad, et al. · arXiv (Cornell University)
PDF
Silenced Biases: The Dark Side LLMs Learned to Refuse
2026 · Rom Himelstein, Amit LeVi, et al. · Proceedings of the AAAI Conference on Artificial Intelligence
PDF
Silenced Biases: The Dark Side LLMs Learned to Refuse
2026 · Association for Artificial Intelligence 2026, Himelstein, Rom, et al. · Underline Science Inc.
1165 - Silenced Biases: The Dark Side LLMs Learned to Refuse
2026 · Association for Artificial Intelligence 2026, Himelstein, Rom, et al. · Underline Science Inc.
Works count: 9
Citation count: 23
H-index: 3
I10-index: 1
Year
Topic
Type