Denis Kleyko

Örebro University

Papers on Lattice

Total citations

Topics

Publication activitypapers/week, last 8 weeks

Research focus

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

Frequent co-authors

Matteo Gioele Collu (1)Riccardo Conte (1)Alberto Giaretta (1)Matteo Zavatteri (1)

Papers (1)

May 27, 2026

University of PaduaMay 27, 2026·also Fondazione Bruno Kessler, Örebro

Refusal Before Decoding: Detecting and Exploiting Refusal Signals in Intermediate LLM Activations

LLMs telegraph their refusals internally, letting you cut adversarial prompt search time by up to 72% by listening to their "gut feelings" before they even speak.

Matteo Gioele Collu, Riccardo Conte, Alberto Giaretta +3

Constitutional AI & AI Ethics Interpretability & Mechanistic Interp Red-Teaming & Adversarial Robustness

Search

Denis Kleyko

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)