Cross-Prompt Reuse of Entity-Replacement Responses
Read the original on arXiv Computation and Language →The paper investigates how entity‑replacement responses can be reused across different English factual questions when model weights are fixed. By injecting hidden‑state changes from two source questions into a target question about the same entity, the authors compare predicted log‑probability shifts with actual measurements, finding that hidden reuse can reduce no‑change error by up to 49% on certain models and that a later protocol can achieve 58‑68% success on three cohorts. The study also notes that larger checkpoints, chat formats, and new entities affect replication, and that relation‑neutral changes at entity tokens lower relation‑selective error more effectively than random changes.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.