StatLLaMA：面向统计领域优化的多阶段训练方法
StatLLaMA: Multi-Stage training for domain-optimized statistical large language models
AI总结：
StatLLaMA通过多阶段训练方法优化统计领域，实现高效且平衡的模型性能。
摘要：
本研究探讨如何高效地构建一个针对统计领域的大型语言模型（LLM）。我们系统比较了三种多阶段训练流程：从无指令跟随能力的基础模型（FM）开始，从增强后指令微调的基础模型（FM）开始，以及从具有强一般推理能力的指令微调模型（FM）开始。这些流程涵盖了持续预训练、监督微调（SFT）、人类反馈强化学习（RLHF）偏好对齐以及下游任务微调（DTFT）。结果显示，从基础模型开始的流程即使经过大量指令微调、SFT或RLHF对齐，也难以发展出有意义的统计推理能力。相反，从LLaMA-3.2-3B-Instruct开始能够实现有效的领域专业化。对SFT变体的全面评估揭示了领域专业知识与一般推理能力之间的明显权衡。我们进一步证明，直接偏好优化能提供稳定有效的RLHF偏好对齐。最后，我们展示DTFT必须以极低强度进行，以避免在高度优化模型中出现灾难性遗忘。最终模型StatLLaMA在数学推理、常识推理和统计专业知识基准上表现出色，为开发资源高效的统计LLM提供了实用蓝图。代码可在https://github.com/HuangDLab/StatLLaMA获取。
英文摘要：
This study investigates how to efficiently build a domain-specialized large language model (LLM) for statistics using the lightweight LLaMA-3.2-3B family as the foundation model (FM). We systematically compare three multi-stage training pipelines--starting from a base FM with no instruction-following capability, a base FM augmented with post-hoc instruction tuning, and an instruction-tuned FM with strong general reasoning abilities--across continual pretraining, supervised fine-tuning (SFT), reinforcement learning from human feedback (RLHF) preference alignment, and downstream task fine-tuning (DTFT). Results show that pipelines beginning with a base FM fail to develop meaningful statistical reasoning, even after extensive instruction tuning, SFT, or RLHF alignment. In contrast, starting from LLaMA-3.2-3B-Instruct enables effective domain specialization. A comprehensive evaluation of SFT variants reveals clear trade-offs between domain expertise and general reasoning ability. We further demonstrate that direct preference optimization provides stable and effective RLHF preference alignment. Finally, we show that DTFT must be performed with extremely low intensity to avoid catastrophic forgetting in highly optimized models. The final model, StatLLaMA, achieves strong and balanced performance on benchmarks of mathematical reasoning, common-sense reasoning, and statistical expertise, offering a practical blueprint for developing resource-efficient statistical LLMs. The code is available at https://github.com/HuangDLab/StatLLaMA.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
