Research

Paper

AI LLM March 19, 2026

Automatic Configuration of LLM Post-Training Pipelines

Authors

Channe Chwa, Xinle Wu, Yao Lu

Abstract

LLM post-training pipelines that combine supervised fine-tuning and reinforcement learning are difficult to configure under realistic compute budgets: the configuration space is high-dimensional and heterogeneous, stages are strongly coupled, and each end-to-end evaluation is expensive. We propose AutoPipe, a budget-aware two-stage framework for configuration selection in LLM post-training. Offline, AutoPipe learns a dataset-conditioned learning-to-rank surrogate from historical runs, capturing within-dataset preferences and providing transferable guidance toward promising regions of the configuration space. Online, for a new dataset, AutoPipe uses the offline guidance to steer Bayesian optimization and models dataset-specific deviations with a Gaussian-process residual surrogate. To reduce evaluation cost, each trial is early-stopped and scored by a learned predictor that maps early training signals to a low-cost proxy for final post-training performance. Experiments on biomedical reasoning tasks show that AutoPipe consistently outperforms offline-only baselines and achieves comparable performance with the strongest online HPO baselines while using less than 10\% of their computational cost.

Metadata

arXiv ID: 2603.18773
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-19
Fetched: 2026-03-20 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.18773v1</id>\n    <title>Automatic Configuration of LLM Post-Training Pipelines</title>\n    <updated>2026-03-19T11:26:56Z</updated>\n    <link href='https://arxiv.org/abs/2603.18773v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.18773v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>LLM post-training pipelines that combine supervised fine-tuning and reinforcement learning are difficult to configure under realistic compute budgets: the configuration space is high-dimensional and heterogeneous, stages are strongly coupled, and each end-to-end evaluation is expensive. We propose AutoPipe, a budget-aware two-stage framework for configuration selection in LLM post-training. Offline, AutoPipe learns a dataset-conditioned learning-to-rank surrogate from historical runs, capturing within-dataset preferences and providing transferable guidance toward promising regions of the configuration space. Online, for a new dataset, AutoPipe uses the offline guidance to steer Bayesian optimization and models dataset-specific deviations with a Gaussian-process residual surrogate. To reduce evaluation cost, each trial is early-stopped and scored by a learned predictor that maps early training signals to a low-cost proxy for final post-training performance. Experiments on biomedical reasoning tasks show that AutoPipe consistently outperforms offline-only baselines and achieves comparable performance with the strongest online HPO baselines while using less than 10\\% of their computational cost.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-19T11:26:56Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Channe Chwa</name>\n    </author>\n    <author>\n      <name>Xinle Wu</name>\n    </author>\n    <author>\n      <name>Yao Lu</name>\n    </author>\n  </entry>"
}