Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Lav R. Varshney

Lav R. Varshney

3 indexed papers

Recent (6 mo)
3
With code
0
Influential cites
0
Benchmarked
0

Publications per year

3
26

Top categories

Crypto×3AI×2ML×1Info Theory×1Software Eng.×1

Frequent co-authors

Benjamin D. Kim1×
Daniel Alabi1×
Royce Moon1×
Max Hartman1×
Vidhata Jayaraman1×
Moulik Choraria1×

Research Timeline

2026
Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models

The paper introduces TraceGuard, a detectability-aware antidistillation method that identifies and poisons 'thought anchors'—sparsely critical sentences—to degrade student model learning without making the defense obvious.

Containment Verification: AI Safety Guarantees Independent of Alignment

The paper introduces containment verification, a novel method that provides safety guarantees by formally verifying the agentic framework itself, ensuring safety regardless of the underlying AI model's capabilities.

Optimal Guarantees for Auditing Rényi Differentially Private Machine Learning

The paper introduces an optimal black-box auditing framework using Donsker-Varadhan estimators to estimate Rényi differential privacy (RDP) guarantees for machine learning algorithms.

Highlighted terms show continued research focus across papers

Papers

cs.LGcs.CRcs.ITRecentMay 21, 2026

Optimal Guarantees for Auditing Rényi Differentially Private Machine Learning

Benjamin D. Kim, Lav R. Varshney, Daniel Alabi

The paper introduces an optimal black-box auditing framework using Donsker-Varadhan estimators to estimate Rényi differential privacy (RDP) guarantees for machine learning algorithms.

View →
cs.AIcs.CRcs.SERecentMay 9, 2026

Containment Verification: AI Safety Guarantees Independent of Alignment

Royce Moon, Lav R. Varshney

The paper introduces containment verification, a novel method that provides safety guarantees by formally verifying the agentic framework itself, ensuring safety regardless of the underlying AI model'…

View →
cs.CRcs.AIRecentApr 25, 2026

Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models

Max Hartman, Vidhata Jayaraman, Moulik Choraria, Yash Savani +1 more

The paper introduces TraceGuard, a detectability-aware antidistillation method that identifies and poisons 'thought anchors'—sparsely critical sentences—to degrade student model learning without makin…

View →