Lift Yourself Up: Retrieval-augmented Text Generation with Self Memory
https://papers.cool/arxiv/2305.02437
Authors: Xin Cheng ; Di Luo ; Xiuying Chen ; Lemao Liu ; Dongyan Zhao ; Rui Yan
Summary: With direct access to human-written reference as memory, retrieval-augmented generation has achieved much progress in a wide range of text generation tasks. Since better memory would typically prompt better generation~(we define this as primal problem). The traditional approach for memory retrieval involves selecting memory that exhibits the highest similarity to the input. However, this method is constrained by the quality of the fixed corpus from which memory is retrieved. In this paper, by exploring the duality of the primal problem: better generation also prompts better memory, we propose a novel framework, selfmem, which addresses this limitation by iteratively employing a retrieval-augmented generator to create an unbounded memory pool and using a memory selector to choose one output as memory for the subsequent generation round. This enables the model to leverage its own output, referred to as self-memory, for improved generation. We evaluate the effectiveness of selfmem on three distinct text generation tasks: neural machine translation, abstractive text summarization, and dialogue generation, under two generation paradigms: fine-tuned small model and few-shot LLM. Our approach achieves state-of-the-art results in four directions in JRC-Acquis, XSum (50.3 ROUGE-1), and BigPatent (62.9 ROUGE-1), demonstrating the potential of self-memory in enhancing retrieval-augmented generation models. Furthermore, we conduct thorough analyses of each component in the selfmem framework to identify bottlenecks and provide insights for future research.
Q: 这篇论文试图解决什么问题?
A: 这篇论文试图解决的问题是现有检索增强型文本生成模型中存在的一个根本性限制:记忆(memory)的有限性。在传统的检索增强型文本生成任务中,模型通常从一个固定的语料库中检索与输入文本最相似的记忆(memory),然后基于这些记忆生成目标文本。然而,这种方法受限于固定语料库的质量,导致生成模型的潜力受到限制。
为了克服这一限制,论文提出了一个新的框架——Selfmem,它通过迭代地使用检索增强型生成器创建一个无界的内存池,并使用内存选择器(memory selector)从生成的候选池中选择一个输出作为下一轮生成的记忆。这种方法允许模型利用自己的输出(即自记忆,self-memory)来改进生成过程。通过结合生成(primal problem)和记忆选择(dual problem),Selfmem框架能够通过模型自身的输出提升自己,从而在多个文本生成任务中实现更好的性能。
Q: 有哪些相关研究?
A: 相关研究主要集中在以下几个领域:
这些相关研究为Selfmem框架提供了理论基础和技术支持,同时也展示了文本生成领域在检索增强、重排序、大型模型应用和评估方法等方面的广泛探索。
Q: 论文如何解决这个问题?
A: 论文提出了一个名为Selfmem的框架来解决检索增强型文本生成模型中记忆有限性的问题。Selfmem框架的核心思想是利用模型自身的输出(自记忆)来提升生成质量。具体来说,Selfmem框架包含两个主要组件: