RSS

Evgenii Kortukov

Karma: 20

Aspiring AI safety researcher. Currently doing my PhD at Fraunhofer HHI in Berlin, focusing on LLM interpretability. Interested in the internal structure underlying safety-relevant behaviors in LLMs: prompt injections, jailbreaks, deception.

A Be­havi­oural and Rep­re­sen­ta­tional Eval­u­a­tion of Goal-di­rect­ed­ness in Lan­guage Model Agents

Gabriele Sarti, Raghu Arghal, ndalton, Fade Chen, Evgenii Kortukov, Calum McNamara, Angelos Nalmpantis, Moksh Nirvaan and Mario Giulianelli
5 Mar 2026 1:08 UTC
20 points
0 comments7 min readLW link

Model­ling, Mea­sur­ing, and In­ter­ven­ing on Goal-di­rected Be­havi­our in AI Systems

Mario Giulianelli, Raghu Arghal, Fade Chen, ndalton, Evgenii Kortukov, Calum McNamara, Angelos Nalmpantis, Moksh Nirvaan and Gabriele Sarti
31 Oct 2025 1:28 UTC
15 points
0 comments8 min readLW link