arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.08526cs.LGcs.CL
收藏

LLM能否在不遗忘的情况下增量学习新概念?

Can LLMs Learn New Concepts Incrementally without Forgetting?

School of Computer Science and Engineering(计算机科学与工程学院);South China University of Technology(华南理工大学)

Junhao Zheng, Shengjie Qiu, Qianli Ma

AI总结:

本文构建含1,023个新概念的Concept-1K基准,评估LLM增量学习能力,发现其仍存在灾难性遗忘,并分析LoRA、规模等因素影响。

摘要:

大语言模型(LLM)已在各类任务上取得显著成功,但其在不遗忘情况下进行增量学习的能力仍未得到充分研究。增量学习(IL)至关重要,因为它使模型能够在获取新知识的同时保留先前学到的信息,类似于人类学习。由于数据泄露问题以及LLM能力过强,现有IL基准并不充分。为应对这些挑战,我们提出Concept-1K,这是一个包含跨多个领域的1,023个近期出现概念的新型数据集。Concept-1K中的概念是离散、可解释的知识单元,可对学习和遗忘过程进行细粒度分析。以Concept-1K作为测试平台,我们旨在回答:“LLM能否像人类一样在不遗忘的情况下增量学习新概念?”研究表明,LLM仍遭受灾难性遗忘,而LoRA尽管微调参数更少,却可能在训练数据上导致更多遗忘。此外,我们还探究了上下文学习、模型规模、缓冲区大小和预训练对IL性能的作用。这些发现揭示了LLM在IL场景中的优势与局限,并为未来研究提供了稳健基准。

英文摘要:

Large Language Models (LLMs) have achieved remarkable success across various tasks, yet their ability to learn incrementally without forgetting remains underexplored. Incremental learning (IL) is crucial as it enables models to acquire new knowledge while retaining previously learned information, akin to human learning. Existing benchmarks for IL are insufficient due to data leakage issues and the overqualification of LLMs. To address these challenges, we introduce Concept-1K, a novel dataset comprising 1,023 recently emerged concepts across diverse domains. The concepts in Concept-1K are discrete, interpretable units of knowledge that allow for fine-grained analysis of learning and forgetting processes. Using Concept-1K as a testbed, we aim to answer the question: ``Can LLMs learn new concepts incrementally without forgetting like humans?'' Our investigation reveals that LLMs still suffer from catastrophic forgetting and that LoRA, despite fine-tuning fewer parameters, may lead to more forgetting on training data. Additionally, we explore the roles of in-context learning, model scale, buffer size, and pretraining in IL performance. These findings highlight the strengths and limitations of LLMs in IL scenarios and provide a robust benchmark for future research.

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。

↑