揭秘面向金融大语言模型的领域自适应后训练
Demystifying Domain-adaptive Post-training for Financial LLMs
AI总结：
针对金融领域大语言模型领域自适应后训练缺乏最优策略的问题，提出包含能力定义、训练方案、数据集与评估套件的FINDAP框架，产出的Llama-Fin模型在多项金融任务上达SOTA，为领域适配提供参考。
摘要：
大语言模型（LLM）的领域自适应后训练已成为医学、金融等专业领域中颇具前景的方法。然而，在不同数据与模型配置下确定最优适配准则和训练策略仍存在重大挑战。为应对这些挑战，我们提出FINDAP——一项针对金融领域LLM领域自适应后训练的系统性细粒度研究。该方法包含四大核心组件：FinCap，用于定义目标领域所需的核心能力；FinRec，一种联合优化持续预训练与指令跟随的高效训练方案，同时搭配一种利用生成式奖励模型过程信号的新型偏好数据蒸馏方法；FinTrain，一套支撑FinRec的精选训练数据集；以及FinEval，一套与FinCap对齐的综合评估套件。最终得到的模型Llama-Fin在广泛的金融任务上取得了当前最优性能。我们的分析还揭示了后训练各阶段对不同能力的贡献，探明了具体挑战与有效解决方案，为LLM的领域适配提供了宝贵洞见。
英文摘要：
Domain-adaptive post-training of large language models (LLMs) has emerged as a promising approach for specialized domains such as medicine and finance. However, significant challenges remain in identifying optimal adaptation criteria and training strategies across varying data and model configurations. To address these challenges, we introduce FINDAP, a systematic and fine-grained investigation into domain-adaptive post-training of LLMs for the finance domain. Our approach consists of four key components: FinCap, which defines the core capabilities required for the target domain; FinRec, an effective training recipe that jointly optimizes continual pre-training and instruction-following, along with a novel preference data distillation method leveraging process signals from a generative reward model; FinTrain, a curated set of training datasets supporting FinRec; and FinEval, a comprehensive evaluation suite aligned with FinCap. The resulting model, Llama-Fin, achieves state-of-the-art performance across a wide range of financial tasks. Our analysis also highlights how each post-training stage contributes to distinct capabilities, uncovering specific challenges and effective solutions, providing valuable insights for domain adaptation of LLMs
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。
