arXiv:2606. 25086v1 Announce Type: new Abstract: Many modern Language Model (LM) pipelines return an averaged model, such as an exponential moving average of the training iterates, rather than the final iterate itself.
arXiv:2602.03702v2 Announce Type: replace-cross Abstract: Large language models are increasingly trained in continual or open-ended settings, where the total training horizon is not known in advance....
arXiv:2510. 14717v2 Announce Type: replace-cross Abstract: Increasing the batch size during training -- a ''batch ramp'' -- is a promising strategy to accelerate large language model pretraining.
arXiv:2603. 09923v4 Announce Type: replace Abstract: Exponential moving averages (EMAs) are a central component of widely used adaptive optimizers such as Adam.
arXiv:2606. 03938v1 Announce Type: cross Abstract: Multi-epoch training is becoming the standard now that compute is growing faster than the supply of high-quality text.
arXiv:2609.36945v1 Announce Type: new Abstract: We study the learning dynamics of fine-tuning a policy model on self-generated and reward-weighted data, with particular focus on a generalized version...
arXiv:2608.29296v1 Announce Type: cross Abstract: Larger batches reduce the variance of stochastic gradients per update and are therefore often expected to accelerate training. Yet whether this stati...
arXiv:2609.25482v1 Announce Type: new Abstract: Large language model (LLM) pretraining conventionally returns the raw final iterate. This couples two design choices: the learning-rate schedule that g...
arXiv:2606. 17526v1 Announce Type: new Abstract: Efficient optimization is essential for training large language models.
arXiv:2607. 10959v1 Announce Type: new Abstract: Standard learning rate schedules such as cosine annealing are tied to a fixed training horizon, limiting their ability to accommodate post hoc horizon extension.
arXiv:2601. 18510v2 Announce Type: replace-cross Abstract: While Large Language Model (LLM) agents excel at general tasks, they inherently struggle with continual adaptation due to the frozen weights after deployment.
arXiv:2607. 21716v1 Announce Type: new Abstract: Due to the complexity of neural network loss landscapes, optimization theory is forced to rely on idealized models, and there is generally a tradeoff between how theoretically tractable the model is, and how accurately it describes the true optimization dynamics.
arXiv:2608. 01418v1 Announce Type: cross Abstract: Autoregressive rollout generation is a major computational cost in reinforcement learning for large language models.