الکس دیماکیس یه پست خفن گذاشته و رفته سراغ یه مسئله‌ی ۲۵ ساله تو حوزه ارتباطات بی‌سیم که قدیما مخ دانشمندا رو به کار گرفته بود، اما حالا مدل‌های جدید هوش مصنوعی حلش کردن

This writeup is fantastic. Not so much because it solves a problem that stumped me and many of my students for many years. But because it shows a new process that I think many in theory research are learning now: Super-intelligence is coming, so complexity in proofs is no longer the bottleneck. Verification of correctness, elementary steps and good storytelling is. I particularly enjoyed this part: ‘…But there was yet another problem, this new proof was UNREADABLE: a wall of notation, variables pointing to variables pointing to ratios of variables defining other variables, exotic matrix-analysis and probability machinery, Marchenko–Pastur adjacent stuff that gives me the hives, and other beautiful things. So for around 4-5 days I kept going back and forth between the two models and asking them to give me the dumbest possible set of steps, for each of the big components needed for the proof to work out.’ It used to be that to get published in fancy theory venues it helped if your proof used fancy tools: exotic matrix-analysis and Marchenko–Pastur etc. This is no longer the case.