Rethinking Easy-to-Hard: Limits of Curriculum Learning in Post-Training for Deductive Reasoning

• URL: http://arxiv.org/abs/2603.27226v1
• Date: Sat, 28 Mar 2026 10:43:36 GMT
• Title: Rethinking Easy-to-Hard: Limits of Curriculum Learning in Post-Training for Deductive Reasoning
• Abstract summary: 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.
• Score: 61.81622271655058
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Curriculum learning (CL), motivated by the intuition that learning in increasing order of difficulty should ease generalization, is commonly adopted both in pre-training and post-training of large language models (LLMs). The intuition of CL is particularly compelling for compositional reasoning, where complex problems are built from elementary inference rules; however, the actual impact of CL on such tasks remains largely underexplored. We present a systematic empirical study of CL for post-training of LLMs, using synthetic arithmetic and logical benchmarks where difficulty is characterized by reasoning complexity rather than surface-level proxies. Surprisingly, across multiple model families and curriculum schedules, we find no robust advantage in difficulty-based sequencing over standard random sampling in either accuracy or response length. These findings persist across both supervised fine-tuning (SFT) and reinforcement learning (RL) methods. Our study suggests that, in the context of deductive reasoning, the specific ordering of training examples plays a negligible role in achieving compositional generalization, challenging the practical utility of curriculum-based post-training.

Related papers

• Many-Shot CoT-ICL: Making In-Context Learning Truly Learn [58.439517684779936]
  In-context learning (ICL) adapts large language models to new tasks by conditioning on demonstrations in the prompt without parameter updates.<n>We study many-shot chain-of-thought in-context learning (CoT-ICL) for reasoning and show that standard many-shot rules do not transfer.
  arXiv  Detail & Related papers  (2026-05-13T13:30:12Z)
• How and Why LLMs Generalize: A Fine-Grained Analysis of LLM Reasoning from Cognitive Behaviors to Low-Level Patterns [51.02752099869218]
  Large Language Models (LLMs) display strikingly different generalization behaviors.<n>We introduce a novel benchmark that decomposes reasoning into atomic core skills.<n>We show that RL-tuned models maintain more stable behavioral profiles and resist collapse in reasoning skills, whereas SFT models exhibit sharper drift and overfit to surface patterns.
  arXiv  Detail & Related papers  (2025-12-30T08:16:20Z)
• 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)
• Iterative In-Context Learning to Enhance LLMs Abstract Reasoning: The Case-Study of Algebraic Tasks [40.48180253367968]
  We introduce an in-context learning methodology that improves the generalization capabilities of general purpose LLMs.<n>Our approach employs an iterative example selection strategy, which incrementally constructs a tailored set of few-shot examples.<n>Our experiments reveal that some LLMs achieve better generalization performances when prompted with simpler few-shot examples.
  arXiv  Detail & Related papers  (2025-09-01T08:54:45Z)
• LogicPuzzleRL: Cultivating Robust Mathematical Reasoning in LLMs via Reinforcement Learning [29.047063129464494]
  Large language models (LLMs) excel at many supervised tasks but often struggle with structured reasoning unfamiliar settings.<n>This discrepancy suggests that standard fine-tuning pipelines may instill narrow, domain-specifics rather than fostering general-purpose thinking strategies.<n>We propose a "play to learn" framework that fine-tunes LLMs through reinforcement learning on a suite of seven custom logic puzzles.
  arXiv  Detail & Related papers  (2025-06-05T09:40:47Z)
• Does the Definition of Difficulty Matter? Scoring Functions and their Role for Curriculum Learning [42.4526628515253]
  Curriculum learning (CL) describes a machine learning training strategy in which samples are gradually introduced into the training process based on their difficulty. We study the robustness and similarity of the most common scoring functions for sample difficulty estimation. We find that the robustness of scoring functions across random seeds positively correlates with CL performance.
  arXiv  Detail & Related papers  (2024-11-01T18:55:31Z)
• Let's Learn Step by Step: Enhancing In-Context Learning Ability with Curriculum Learning [9.660673938961416]
  Demonstration ordering is an important strategy for in-context learning (ICL) We propose a simple but effective demonstration ordering method for ICL, named the few-shot In-Context Curriculum Learning (ICCL)
  arXiv  Detail & Related papers  (2024-02-16T14:55:33Z)
• LaRS: Latent Reasoning Skills for Chain-of-Thought Reasoning [61.7853049843921]
  Chain-of-thought (CoT) prompting is a popular in-context learning approach for large language models (LLMs)<n>This paper introduces a new approach named Latent Reasoning Skills (LaRS) that employs unsupervised learning to create a latent space representation of rationales.
  arXiv  Detail & Related papers  (2023-12-07T20:36:10Z)
• In-context Learning Generalizes, But Not Always Robustly: The Case of Syntax [36.98247762224868]
  In-context learning (ICL) is now a common method for teaching large language models (LLMs) new tasks. Do models infer the underlying structure of the task defined by the context, or do they rely on superficial generalizations that only generalize to identically distributed examples? In experiments with models from the GPT, PaLM, and Llama 2 families, we find large variance across LMs. The variance is explained more by the composition of the pre-training corpus and supervision methods than by model size.
  arXiv  Detail & Related papers  (2023-11-13T23:52:43Z)
• Skills-in-Context Prompting: Unlocking Compositionality in Large Language Models [68.18370230899102]
  We investigate how to elicit compositional generalization capabilities in large language models (LLMs) We find that demonstrating both foundational skills and compositional examples grounded in these skills within the same prompt context is crucial. We show that fine-tuning LLMs with SKiC-style data can elicit zero-shot weak-to-strong generalization.
  arXiv  Detail & Related papers  (2023-08-01T05:54:12Z)

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.