We use perplexity to measure LLMs a lot. What if we instead used how close they were to judging how something would end?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-longterm-predictability-as-2026,
author = {Holtzman, Ari},
title = {Long-term predictability as a metric},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/rcdOxjZLjWrAZlZWo5uP}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!