If we train a small Transformer to simply decide when to copy and from where, will it be more accurate at deciding than training a language model and using it to make the same copying decisions?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-is-copying-more-2025,
author = {Holtzman, Ari},
title = {Is copying more calibrated if it's specialized?},
year = {2025},
url = {https://hypogenic.ai/ideahub/idea/1CZpEFT3Gp2PA8W3QGlk}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!