Paper Detail
Yuda Song, Alvin Zhang, Fahim Tajwar, Daman Arora, Jeff Schneider, Daniel Khashabi, Ruslan Salakhutdinov, J. Andrew Bagnell, Aarti Singh, Andrea Zanette
Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not fundamental. By deriving suitable RL objectives as maximum-likelihood fitting to model-generated target distributions, we place all of these stages under a single likelihood-based training algorithm, unified training. This perspective turns the foundation-model training lifecycle into a problem of choosing which data to train on, in what proportions, and when, with conventional staged training as a special case. We instantiate this framework in a unified training system that mixes static data and RL generations within a single trainer. Across controlled and real-world experiments, unified training improves the accuracy-compute tradeoff and task coverage while preserving broader capabilities, and provides a stronger starting point for continual learning. This work lays the groundwork for a unified science of foundation-model training, connecting mathematical principles, training infrastructure, and the development of model capabilities.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{song2026training,
title = {Training Foundation Models with a Single Algorithm},
author = {Yuda Song and Alvin Zhang and Fahim Tajwar and Daman Arora and Jeff Schneider and Daniel Khashabi and Ruslan Salakhutdinov and J. Andrew Bagnell and Aarti Singh and Andrea Zanette},
year = {2026},
abstract = {Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not fundamental. By deriving suitable RL objectives as maximum-likelihood fitting to model-generated target distributions, we place all of these stages under a single likelihood-based training algorithm, unified training. This perspective turns the foundation-model},
url = {https://arxiv.org/abs/2610.40564},
keywords = {cs.LG, maximum-likelihood fitting, RL objectives, unified training, foundation models, reinforcement learning, continual learning, code available, huggingface daily},
eprint = {2610.40564},
archiveprefix = {arXiv},
}
{}