| title | Cross-Prompt Reuse of Entity-Replacement Responses |
|---|---|
| authors | Jinkun Chen, Xuemin Yu, Sijia Han, Vlado Keselj |
| date | 2026-10-30 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.00352 |
| upvotes | 1 (Hugging Face Daily Papers, 2026-11-05) |
| links | abs pdf hf |
| cite | |
| abstract | With model weights fixed, we measure how much of an entity-replacement response transfers between English factual questions, beyond whether answers change. We add the hidden-state change from two source questions to a target question about the same entity, comparing the predicted log-probability shift between two relations' published answers, given known answer prefixes, with the measured one. Hidden reuse injects the aligned change at 50% depth before the target's remaining layers; direct reuse adds source output-score changes. With other-relation sources, hidden reuse removes 45% and 49% of no-change error on 0.5-1B Qwen and Llama. Gemma-2-2B removes 22% and passes one of three comparisons in plain text without its default beginning-of-sequence (BOS) token. A later protocol restoring it yields 58-68%, passing every comparison on three cohorts. Larger checkpoints, chat format and new entities show replication and format- or cohort-dependent failures. On small plain-format checkpoints, other-relation hidden reuse mainly reproduces the shift both relations' questions share, without reliably lowering relation-selective error. With target-relation sources, hidden reuse's post hoc advantage over direct reuse is 0.2-0.9 nats outside chat format, unconfirmed and not replacement-specific. The same relation-neutral change at both questions' entity tokens lowers relation-selective error more than random or other-replacement changes, indicating receiving-question dependence. |
bibtex
@misc{chen2026cross,
title={Cross-Prompt Reuse of Entity-Replacement Responses},
author={Jinkun Chen and Xuemin Yu and Sijia Han and Vlado Keselj},
year={2026},
eprint={2611.00352},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.00352}
}