RA-ISF：通过迭代自反馈从检索增强中学习回答与理解
RA-ISF: Learning to Answer and Understand from Retrieval Augmentation via Iterative Self-Feedback
AI总结：
本文提出RA-ISF框架，通过检索增强与迭代自反馈分解任务并分三个子模块处理，在GPT3.5、Llama2等模型上提升事实推理能力并减少幻觉。
摘要：
大型语言模型（LLMs）在众多任务中表现出卓越性能，但仍严重依赖存储在其参数中的知识。此外，更新这些知识会产生高昂的训练成本。检索增强生成（RAG）方法通过整合外部知识来解决这一问题。模型可以通过检索与查询相关的知识来回答以前无法回答的问题。这种方法在特定任务的某些场景中提高了性能。然而，如果检索到不相关的文本，可能会损害模型性能。在本文中，我们提出了检索增强迭代自反馈（RA-ISF），这是一个通过迭代分解任务并在三个子模块中处理任务以增强模型问题解决能力的框架。实验表明，我们的方法优于现有基准，在GPT3.5、Llama2等模型上表现良好，显著增强了事实推理能力并减少了幻觉。
英文摘要：
Large language models (LLMs) demonstrate exceptional performance in numerous tasks but still heavily rely on knowledge stored in their parameters. Moreover, updating this knowledge incurs high training costs. Retrieval-augmented generation (RAG) methods address this issue by integrating external knowledge. The model can answer questions it couldn't previously by retrieving knowledge relevant to the query. This approach improves performance in certain scenarios for specific tasks. However, if irrelevant texts are retrieved, it may impair model performance. In this paper, we propose Retrieval Augmented Iterative Self-Feedback (RA-ISF), a framework that iteratively decomposes tasks and processes them in three submodules to enhance the model's problem-solving capabilities. Experiments show that our method outperforms existing benchmarks, performing well on models like GPT3.5, Llama2, significantly enhancing factual reasoning capabilities and reducing hallucinations.
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。
