Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics

• URL: http://arxiv.org/abs/2609.09099v1
• Date: Tue, 08 Sep 2026 17:39:16 GMT
• Title: Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics
• Abstract summary: We present Wasserstein curriculum paths, a simple framework that decouples factors by representing curricula as trajectories of training distributions over discrete difficulty levels.<n>Across a calibrated synthetic suite with 12 tasks and 33 difficulty axes, we use this framework to isolate the effects of ordering, matched exposure, endpoint smoothness, and pacing under fixed training budgets.
• Score: 11.91720769864227
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Curriculum learning is governed by several coupled design choices---how difficulty is defined, how examples are ordered, how much exposure each level receives, and how quickly training moves across levels---making it hard to isolate what actually helps. We present Wasserstein curriculum paths, a simple transport-based framework that decouples these factors by representing curricula as trajectories of training distributions over discrete difficulty levels. Across a calibrated synthetic suite with 12 tasks and 33 difficulty axes, we use this framework to isolate the effects of ordering, matched exposure, endpoint smoothness, and pacing under fixed training budgets. We find that curriculum effects are strongly context-dependent: no single strategy dominates across tasks, difficulty axes, and budgets, and curricula mainly change where a fixed budget is spent most effectively. Within this framework, easy-to-hard ordering improves hard-level performance relative to exposure-matched static sampling, showing that the benefit is not explained by cumulative exposure alone. We further show that endpoint smoothness and pacing substantially affect where along the difficulty spectrum a curriculum is effective. Finally, we show that the same transport view naturally supports extensions to learned pacing through geometry and to structured difficulty spaces beyond one-dimensional orderings.

Related papers

• Trajectory-Level Automatic Curriculum Learning for Legged Locomotion on Unstructured Terrain [30.327576375647663]
  textbfourname generates training tasks directly from unstructured terrain maps.<n>Our method improves success rate by (56.3%) over direct training without curriculum.
  arXiv  Detail & Related papers  (2026-08-17T06:30:45Z)
• Rethinking Easy-to-Hard: Limits of Curriculum Learning in Post-Training for Deductive Reasoning [61.81622271655058]
  We present a systematic empirical study of curriculum-based post-training of large language models.<n>We find no robust advantage in difficulty-based sequencing over standard random sampling in either accuracy or response length.<n>Our study suggests that, in the context of deductive reasoning, the specific ordering of training examples plays a negligible role in achieving compositional generalization.
  arXiv  Detail & Related papers  (2026-03-28T10:43:36Z)
• Level Up: Defining and Exploiting Transitional Problems for Curriculum Learning [11.76670314650204]
  We introduce a novel method for measuring the difficulty of individual problem instances directly relative to the ability of a given model.<n>We find that training on a curriculum that "levels up" from easier to harder transitional problems most efficiently improves a model to the next tier of competence.
  arXiv  Detail & Related papers  (2026-03-14T05:17:45Z)
• Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training [76.12556589212666]
  We show that curriculum post-training avoids the exponential complexity bottleneck.<n>Under outcome-only reward signals, reinforcement learning finetuning achieves high accuracy with sample complexity.<n>We establish guarantees for test-time scaling, where curriculum-aware querying reduces both reward oracle calls and sampling cost from exponential to order.
  arXiv  Detail & Related papers  (2025-11-10T18:29:54Z)
• What Makes a Good Curriculum? Disentangling the Effects of Data Ordering on LLM Mathematical Reasoning [47.10036442634749]
  Curriculum learning (CL) has become a popular strategy for improving reasoning in large language models (LLMs)<n>Yet prior work employs disparate difficulty metrics and training setups, leaving open fundamental questions: When does curriculum help?<n>We address these questions through a unified offline evaluation framework that decomposes curriculum difficulty into five complementary dimensions: Problem Difficulty, Model Surprisal, Confidence Margin, Predictive Uncertainty, and Decision Variability.
  arXiv  Detail & Related papers  (2025-10-21T21:43:38Z)
• Your Pretrained Model Tells the Difficulty Itself: A Self-Adaptive Curriculum Learning Paradigm for Natural Language Understanding [53.63482987410292]
  We present a self-adaptive curriculum learning paradigm that prioritizes fine-tuning examples based on difficulty scores predicted by pre-trained language models.<n>We evaluate our method on four natural language understanding (NLU) datasets covering both binary and multi-class classification tasks.
  arXiv  Detail & Related papers  (2025-07-13T19:36:17Z)
• When Do Curricula Work in Federated Learning? [56.88941905240137]
  We find that curriculum learning largely alleviates non-IIDness. The more disparate the data distributions across clients the more they benefit from learning. We propose a novel client selection technique that benefits from the real-world disparity in the clients.
  arXiv  Detail & Related papers  (2022-12-24T11:02:35Z)
• Let the Model Decide its Curriculum for Multitask Learning [22.043291547405545]
  We propose two classes of techniques to arrange training instances into a learning curriculum based on difficulty scores computed via model-based approaches. We show that instance-level and dataset-level techniques result in strong representations as they lead to an average performance improvement of 4.17% and 3.15% over their respective baselines.
  arXiv  Detail & Related papers  (2022-05-19T23:34:22Z)
• CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition [79.92240030758575]
  We propose a novel Adaptive Curriculum Learning loss (CurricularFace) that embeds the idea of curriculum learning into the loss function. Our CurricularFace adaptively adjusts the relative importance of easy and hard samples during different training stages.
  arXiv  Detail & Related papers  (2020-04-01T08:43:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.