12 papers · 1 filter
Model Unlearning Objectives Vary for Distinct Language Functions
Berk Atil, Vipul Gupta, Rebecca J. Passonneau
Large language models (LLMs) learn undesirable properties during pretraining, including dangerous knowledge and toxic text generation. Just as post-training uses different objectiv…
Robust Persona-Aware Toxicity Detection with Prompt Optimization and Learned Ensembling
Berk Atil, Rebecca J. Passonneau, Ninareh Mehrabi
Toxicity detection is inherently subjective, shaped by the diverse perspectives and social priors of different demographic groups. While ``pluralistic'' modeling as used in economi…
Something Just Like TRuST : Toxicity Recognition of Span and Target
Berk Atil, Namrata Sureddy, Rebecca J. Passonneau
Toxic language includes content that is offensive, abusive, or that promotes harm. Progress in preventing toxic output from large language models (LLMs) is hampered by inconsistent…
Do Methods to Jailbreak and Defend LLMs Generalize Across Languages?
Berk Atil, Rebecca J. Passonneau, Fred Morstatter
Large language models (LLMs) undergo safety alignment after training and tuning, yet recent work shows that safety can be bypassed through jailbreak attacks. While many jailbreaks…
Can LLMs Rank the Harmfulness of Smaller LLMs? We are Not There Yet
Berk Atil, Vipul Gupta, Sarkar Snigdha Sarathi Das +1
Large language models (LLMs) have become ubiquitous, thus it is important to understand their risks and limitations. Smaller LLMs can be deployed where compute resources are constr…
Non-Determinism of "Deterministic" LLM Settings
Berk Atil, Sarp Aykent, Alexa Chittams +10
LLM (large language model) practitioners commonly notice that outputs can vary for the same inputs under settings expected to be deterministic. Yet the questions of how pervasive t…