LLM privacy presentations and a human-centered security workshop planned for COLM 2026

On October 6, a lab member said students would present work at COLM 2026 on personalization and privacy concerns in LLM agents, along with CIDER, a dataset about privacy-preference boundaries. The member also said they were co-organizing an October 9 workshop on human-centered privacy and security for language models, with student presentations planned on re-identification, anonymization and authorization alignment for LLM agents.