解开自然语言处理中的课程学习：迈向统一分类法
Disentangling Curriculum Learning in NLP: Towards a Unifying Taxonomy
浏览论文内容
中文总结 AI 辅助
自然语言处理中课程学习缺乏难度函数和调度器的原则性说明，本文提出细粒度分类法，区分难度评估与训练调度，用于系统分析CL策略，揭示不可比性问题，支持策略设计等，并推动改进来源区分的评估实践。
中文摘要 AI 辅助
尽管自然语言处理领域对课程学习（CL）已有十多年的研究，但对于给定问题应使用哪种难度函数或调度器仍缺乏原则性的说明。为了解阻碍这方面进展的因素，我们提出了一种细粒度分类法，将难度评估与训练调度分开，以便对CL策略进行系统分析。对于难度评估，区分了归因来源和任务依赖性，揭示难度是一个视角概念，编码了关于实例难以学习原因的不同假设。对于调度，首次根据预期训练贡献对CL调度器进行形式化，通过引入保留机制和单调性属性实现不同实现之间的比较。应用于对自然语言处理中CL工作的专门分析时，我们的分类法揭示了一个系统性的不可比性问题：先前的工作混淆了难度和调度的不同概念，常在相同CL标签下追求不同目标，阻碍了比较和连贯证据基础的积累。除了诊断，该分类法还支持CL策略的设计、分析和比较，并推动对观察到的改进来源进行区分的评估实践。
英文摘要
Despite more than a decade of curriculum learning (CL) research in NLP, the field lacks a principled account of which difficulty function or scheduler to use for a given problem. To understand what has hindered progress towards this account, we propose a fine-grained taxonomy separating difficulty evaluation from training scheduling to enable systematic analysis of CL strategies. For difficulty evaluation, we distinguish attribution source and task dependence, revealing difficulty as a perspectival concept encoding different assumptions about what makes an instance hard to learn. For scheduling, we provide the first formalisation of CL schedulers in terms of expected training contribution, enabling comparison across implementations by introducing retention regimes and monotonicity properties. Applied in a dedicated analysis of CL works in NLP, our taxonomy reveals a systematic incomparability problem: prior works conflate distinct notions of difficulty and scheduling, often pursuing different objectives under the same CL label -- hindering comparison and the accumulation of a coherent evidence base. Beyond diagnosis, the taxonomy supports the design, analysis, and comparison of CL strategies, and motivates evaluation practices that disentangle the sources of observed improvement.
发表机构
- University of Bonn（波恩大学）
- Lamarr Institute（拉马尔研究所）
- Fraunhofer IAIS（弗劳恩霍夫人工智能研究所）
机构由 AI 辅助整理，请以论文原文为准。
