AI RESEARCH
Modeling Pathology-Like Behavioral Patterns in Language Models Through Behavioral Fine-Tuning
arXiv CS.CL
•
ArXi:2605.22356v1 Announce Type: new Large language models are increasingly used as computational tools for modeling human-like behavior. We Across two architectures, fine-tuned models show stable, context-general shifts in next-token probability distributions, including increased probability assigned to negative and threat-related interpretations in open-ended language tasks. These effects generalize beyond