无需训练，通过检索增强实现低成本自主知识生成
Retrieval Augmented Learning: A Retrial-based Large Language Model Self-Supervised Learning and Autonomous Knowledge Generation
- 利用检索增强生成构建中间数据模块，实现三阶段知识自动生成
- 在决策平台测试中显著降低幻觉，性能提升且成本极低
- 适合需要低成本、高鲁棒性的自动化决策系统
大语言模型（LLM）预训练阶段缺乏领域特定数据，严重限制其在专业场景中的应用，而后续微调又需大量计算资源。本文提出无奖励自监督学习框架Retrial-Augmented Learning（RAL），无需模型训练即可运行。通过将检索增强生成（RAG）转化为组织中间数据的模块，实现了提出假设、验证假设、生成知识的三阶段自主知识生成。在结合复杂性与领域知识需求的LLM-PySC2决策平台进行评估，实验表明该方法有效减少幻觉，提升决策性能，且代价极低。同时在分布外（OOD）任务、鲁棒性和可迁移性方面表现良好，为决策问题和自主知识生成提供了低成本高效的解决方案。
原文摘要 · Abstract (English)
The lack of domain-specific data in the pre-training of Large Language Models (LLMs) severely limits LLM-based decision systems in specialized applications, while post-training a model in the scenarios requires significant computational resources. In this paper, we present Retrial-Augmented Learning (RAL), a reward-free self-supervised learning framework for LLMs that operates without model training. By developing Retrieval-Augmented Generation (RAG) into a module for organizing intermediate data, we realized a three-stage autonomous knowledge generation of proposing a hypothesis, validating the hypothesis, and generating the knowledge. The method is evaluated in the LLM-PySC2 environment, a representative decision-making platform that combines sufficient complexity with domain-specific knowledge requirements. Experiments demonstrate that the proposed method effectively reduces hallucination by generating and utilizing validated knowledge, and increases decision-making performance at an extremely low cost. Meanwhile, the approach exhibits potential in out-of-distribution(OOD) tasks, robustness, and transferability, making it a cost-friendly but effective solution for decision-making problems and autonomous knowledge generation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。