AF3 paper: just increasing seeds lifted scores from 0.3 to 0.6, says sokrypton
sokrypton · x · 2026-09-18
Adding evidence in the same thread, sokrypton notes the AF3 paper went from 0.3 to 0.6 by simply increasing seeds. It follows that training the next model on confident predictions sampled by chance should improve performance, eventually approaching the level of infinite seeds on the initial model.
Related event: sokrypton explains why TorchFold self-distillation works(2 posts)→
More from Research
- alphaXiv launches "Musolesi number" to measure your academic coauthor distance — mircomusolesi · 2026-09-18
- Anthropic quietly builds Bay Area wet lab to let Claude direct biology experiments — mkheck · 2026-09-18
- Opportunistic AI Imaging Opens a New Window Into the Kidney via Routine Echocardiograms — EricTopol · 2026-09-18
- arXiv One-Strike Policy Could Mean a Lifetime STOC Ban, Researcher Warns — Aaroth · 2026-09-18
- RSI-Exam updates: GPT-6-astra holds #1 at 0.5126, Anthropic's Fable 5.1 debuts at #2 — HuaxiuYaoML · 2026-09-18
- Entropy doesn't track what matters in chain-of-thought, BlackBoxNLP paper finds — gsarti_ · 2026-09-18