AI safety researcher

George Kesidis

Pennsylvania State University

Google Scholar dblp ORCID Homepage

Last 12 months3 papers

Attacks 1

AI safety papers from the last year

On this site: 27 papers

October 2026

LLM Trojans that Increase Response Lengths and Cost
Oct 17AttacksarXiv ↗

June 2026

CSO-LLM: Class Subspace Orthogonalization for Post-Training Backdoor Detection and Trigger Inversion in LLMs
Jun 30arXiv ↗

December 2025

Improving the Sensitivity of Backdoor Detectors via Class Subspace Orthogonalization
Dec 9arXiv ↗

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.