This is a concrete proposal for this leaderboard idea from Ari: https://hypogenic.ai/ideahub/idea/AMJ09TAxgffEYAQW8cCh
This focuses on a particular style of approach. If we collect some human texts from a single person, how do LLMs become better at evading AI detectors as the input size grows?
This can work both for prompting and fine-tuning.
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{tan-scaling-law-for-2026,
author = {Tan, Chenhao},
title = {Scaling Law for Humanizing Texts},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/7YOZHZ5VgHg8rDJPbYCj}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!