四象限难度：一种简单的分类及其限制
Four Quadrants of Difficulty: A Simple Categorisation and its Limits
AI总结：
本文提出四象限难度分类，揭示任务无关与依赖特征的差异，挑战课程学习的常见假设，强调任务依赖难度估计器的重要性。
摘要：
课程学习（CL）旨在通过估计样本的难度并相应地安排样本来提高模型训练的成果。在NLP中，难度通常使用任务无关的语言启发式或人类直觉来近似，隐含地假设这些信号与神经网络模型难以学习的东西相关联。我们提出了一种四象限难度信号的分类——人类与模型、任务无关与任务依赖——并系统地分析它们在自然语言理解数据集上的相互作用。我们发现任务无关的特征行为上基本独立，而只有任务依赖的特征相互关联。这些发现挑战了常见的CL直觉，并突显了需要轻量级、任务依赖的难度估计器的必要性，以更好地反映模型的学习行为。
英文摘要：
Curriculum Learning (CL) aims to improve the outcome of model training by estimating the difficulty of samples and scheduling them accordingly. In NLP, difficulty is commonly approximated using task-agnostic linguistic heuristics or human intuition, implicitly assuming that these signals correlate with what neural models find difficult to learn. We propose a four-quadrant categorisation of difficulty signals -- human vs. model and task-agnostic vs. task-dependent -- and systematically analyse their interactions on a natural language understanding dataset. We find that task-agnostic features behave largely independently and that only task-dependent features align. These findings challenge common CL intuitions and highlight the need for lightweight, task-dependent difficulty estimators that better reflect model learning behaviour.
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。
