Xinyue Liu

@xinyuel.bsky.social

PhD student @sbucompsc; Prev @LITatCMU

So excited to share our new paper — and also my first project in my PhD journey! 🎉 It’s striking to see how finetuning on a simple writing task unlocks verbatim recall of copyrighted books from many other authors not in the finetuning data. So grateful to work with such an amazing group of people! ✨

Tuhin Chakrabarty@tuhinchakr.bsky.social · 4mo ago

🚨Paper on AI & Copyright Courts have credited AI companies' claims that alignment prevents reproducing copyrighted data. What if finetuning on a simple writing task breaks it. Worse: tuning on just one author (e.g., Murakami) unlocks verbatim recall of 30+ other authors' books (up to 90%) (1/n)🧵