2 papers
cs.CL2025
Guard Vector: Beyond English LLM Guardrails with Task-Vector Composition and Streaming-Aware Prefix SFT
Wonhyuk Lee, Youngchol Kim, Yunjin Park +3
We introduce Guard Vector, a safety task vector computed as the parameter difference between a guardrail model (Guard Model) and a same-architecture pretrained language model. Comp…
cs.CL2024
Persona Setting Pitfall: Persistent Outgroup Biases in Large Language Models Arising from Social Identity Adoption
Wenchao Dong, Assem Zhunis, Dongyoung Jeong +3
Drawing parallels between human cognition and artificial intelligence, we explored how large language models (LLMs) internalize identities imposed by targeted prompts. Informed by…