Shireen Kudukkil Manchingal

relAI -Konrad Zuse School of Excellence in Reliable AI

Papers on Lattice

Total citations

Topics

Publication activitypapers/week, last 8 weeks

Research focus

Interpretability & Mechanistic Interp (1)Red-Teaming & Adversarial Robustness (1)

Frequent co-authors

S. Nikolenko (1)Sofiia Nikolenko (1)Michele Papucci (1)Mina Rezaei (1)

Papers (1)

Jun 23, 2026

S. Nikolenko +52w ago·also LMU, relAI -Konrad Zuse School of Excellence, Università di Pisa

What Intermediate Layers Know: Detecting Jailbreaks from Entropy Dynamics

Entropy dynamics in intermediate layers reveal that jailbreak behavior is encoded more distinctly than previously understood, challenging conventional defenses.

S. Nikolenko, Sofiia Nikolenko, Michele Papucci +3

Interpretability & Mechanistic Interp Red-Teaming & Adversarial Robustness

Search

Shireen Kudukkil Manchingal

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)