arXiv · 2611.20430
TrainExplorer: Autonomous LLM Post-Training via Plan-and-Decouple Exploration
Abstract
LLM-based agents can automate model post-training, accelerating progress toward recursive self-improvement. Despite their promise, existing approaches remain under-explored and face two main challenges: (1) strategy lock-in that limits exploration beyond the initial strategy; and (2) coupled execution that obscures the effects of individual modifications. To address these challanges, we propose TrainExplorer, a multi-agent framework for autonomous post-training, that incorporates strategy planning and execution decoupling, to achieve broaden strategy exploration. First, strategy planning draws on scientific literature to construct a candidate search space of training strategies. Then, decoupled execution enables single-component interventions and attribution to guide strategy exploration. Extensive experiments on PostTrainBench demonstrate that TrainExplorer achieves a weighted aggregate score of 39.02, outperforming Claude Code using the same LLM by 7.32 and surpassing GPT-5.6 Sol-powered Codex, which scores 36.23. Code will be available.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Xuannan Liu, Tian Lan, Zekun Li, Xing Cui, Zhengfeng Yu, Feihu Jiang, Qianghuai Jia, Longyue Wang, Zhao Xu, Weihua Luo, Peipei Li, Ran He. 2026-11-16. TrainExplorer: Autonomous LLM Post-Training via Plan-and-Decouple Exploration. https://searcharxiv.com/abs/2611.20430
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.