If an LM requires less finetuning data to believe X than Y, is it also more susceptible to believing X in a conversation before finetuning?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-already-present-beliefs-2026,
author = {Holtzman, Ari},
title = {Already Present Beliefs?},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/POf2Qh3nT3UCo6wq2k8l}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!