From Training Exposure to Membership Evidence: A Controlled Study of OLMo 2
Whether a language model was trained on a document matters for copyright, privacy, and benchmark contamination. Yet most membership-inference (MI) evaluations on language models collect members and non-members after the fact, so success may reflect distribution shift and failure is inconclusive. We re-ran OLMo 2 1B's o...
A. Banaei, Muhammad Shahzad, Gerald Q. Maguire et al.
· 0 citations