LLMs must represent what different people find dramatic. What one person X find dramatic and another person Y finds melodramatic are probably related, but they are probably not modeled well as thresholds on a single axis. How do models represent this? A low-dimensional space of drama-features? A low-dimensional space of reader features that contain drama subspaces?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-different-kinds-of-2026,
author = {Holtzman, Ari},
title = {Different Kinds of Drama},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/QHGT7WVSB43EGwceotfR}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!