AI RESEARCH
Just as Humans Need Vaccines, So Do Models: Model Immunization to Combat Falsehoods
arXiv CS.CL
•
ArXi:2505.17870v3 Announce Type: replace Large language models (LLMs) reproduce misinformation not by memorizing false facts alone, but by learning the linguistic patterns that make falsehoods persuasive, such as hedging, false presuppositions, and fabricated citations. We propose model immunization, a