Trojan Detection in Large Language Models: Insights from The Trojan Detection Challenge

no code implementations • 21 Apr 2024 • Narek Maloyan, Ekansh Verma, Bulat Nutfullin, Bislan Ashinov

The phenomenon of unintended triggers and the difficulty in distinguishing them from intended triggers highlights the need for further research into the robustness and interpretability of LLMs.

Uncertainty-Aware Evaluation for Vision-Language Models

1 code implementation • 22 Feb 2024 • Vasily Kostumov, Bulat Nutfullin, Oleg Pilipenko, Eugene Ilyushin

Specifically, we show that models with the highest accuracy may also have the highest uncertainty, which confirms the importance of measuring it for VLMs.

Conformal Prediction Language Modelling +4

DIALOG-22 RuATD Generated Text Detection

1 code implementation • 16 Jun 2022 • Narek Maloyan, Bulat Nutfullin, Eugene Ilyushin

Text Generation Models (TGMs) succeed in creating text that matches human language style reasonably well.

Binary Classification Classification +2