🔮 Wow, revolutionary insight: if models change, evaluations might not work! Who would've thunk it?! 🚀 Let's all panic about the inevitable #doom of #evals and ignore the fact that predicting the future isn't our strong suit – especially when we can't even predict the end of this sentence...
https://wanglun1996.github.io/blog/your-evals-will-break.html #revolutionaryinsight #predictions #futureuncertainty #techhumor #HackerNews #ngated