5 papers
Controllable and explainable personality sliders for LLMs at inference time
Florian Hoppe, David Khachaturov, Robert Mullins +1
Aligning Large Language Models (LLMs) with specific personas typically relies on expensive and monolithic Supervised Fine-Tuning (SFT) or RLHF. While effective, these methods requi…
Tiered Anonymity on Social-Media Platforms as a Countermeasure against Deepfakes and LLM-Driven Mass Misinformation
David Khachaturov, Roxanne Schnyder, Robert Mullins
We argue that governments should mandate a three-tier anonymity framework on social-media platforms as a reactionary measure prompted by the ease-of-production of deepfakes and lar…
"I am bad": Interpreting Stealthy, Universal and Robust Audio Jailbreaks in Audio-Language Models
Isha Gupta, David Khachaturov, Robert Mullins
The rise of multimodal large language models has introduced innovative human-machine interaction paradigms but also significant challenges in machine learning safety. Audio-Languag…
Adversarial Suffix Filtering: a Defense Pipeline for LLMs
David Khachaturov, Robert Mullins
Large Language Models (LLMs) are increasingly embedded in autonomous systems and public-facing environments, yet they remain susceptible to jailbreak vulnerabilities that may under…
Watermarking Needs Input Repetition Masking
David Khachaturov, Robert Mullins, Ilia Shumailov +1
Recent advancements in Large Language Models (LLMs) raised concerns over potential misuse, such as for spreading misinformation. In response two counter measures emerged: machine l…