arXiv:2606. 24993v1 Announce Type: new Abstract: Sequential learning is order-dependent: from Pile-style next-token domain adaptation to instruction-SFT and DPO, N candidate sources induce N!
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.
arXiv:2606. 25178v2 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) has been extended from single-domain training to multi-domain reasoning suites spanning mathematics, programming, and science.
arXiv:2603. 21180v4 Announce Type: replace Abstract: Sequential experimental design under expensive, gradient-free objectives is a central challenge in computational statistics: evaluation budgets are tightly constrained and information must be extracted efficiently from each observation.
arXiv:2608. 03316v1 Announce Type: new Abstract: On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid.
arXiv:2606. 29476v1 Announce Type: cross Abstract: Self-distilled agentic reinforcement learning augments trajectory-level reward with a token-level distillation loss, using as its teacher the same policy conditioned on privileged context.
On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid. We ask what happens when none of this holds, as when the strongest teacher available and the student one wishes to deploy come from different model families, and find that the standard recipes have no answer: teacher latents cannot serve as targets in a foreign coordinate system, per-pixel losses against a teacher that stochastically re-draws local detail degenerate into blur or divergence, and timestep indices lose their meaning across mismatched schedules.
arXiv:2606. 23740v1 Announce Type: cross Abstract: Offline reinforcement-learning losses (RFT, RIFT, DFT, Offline GRPO, DPO) are widely used to distill reasoning from large teachers into smaller students, and are typically compared on downstream accuracy alone.