statement
The post's author concludes that in supervised learning a distribution shift between train and test is usually a problem, while for RL training it is part of the solution.
A statement by כותב הפוסט. Statements are as reported by the cited publishers — not verified by DecisionGateway.
אז מתברר כי בלמידת Supervised קיום של Distribution Shift בין טריין לטסט הוא בדרך כלל בעיה. ועבור אימון RL הוא מהווה חלק מהפתרון.
Discussion
Related claims
The paper proves that when training reasoning models with RL, it is not always true that training data should match the test-time distribution; sometimes data different from what the model will see at test time yields much better test-time results.המאמר · investigationDiffusion training is relatively simple regression — 'here is a noisy image (letter), guess the noise' — with no need for negative examples and no normalization constant to compute, which is why it is stable and scalable.post author · statement'The end of diffusion' is hype; the better question is whether it is worth paying for more expensive training to get built-in verification and flexible compute at inference time — a real and interesting trade-off.מייק · statementPre-training is complete and 4.8 is now in the RL stage, with evaluation scores starting to rise.马斯克 · statementThe researchers show that training leads the model to learn an efficient solution path, even though it receives no extra reward for choosing that path.החוקרים · investigationThe models trained on the easier examples are tested on problems harder than those they saw in post-training and still succeed much more than models trained directly on the test difficulty level, without receiving full solutions or feedback on intermediate steps.המחברים · investigation