arXiv AI October 12, 2026 By Ronghan Li, Fei Chen, Mingyu Lu

A Step Harder, Not a Step Too Far: The Benefits and Limits of Curriculum Difficulty Progression in GRPO

The paper investigates how curriculum difficulty progression affects training in Group Relative Policy Optimization (GRPO). It finds that continually selecting medium‑difficulty samples is not optimal; a staged curriculum that gradually shifts toward harder samples improves performance, reaching a peak accuracy of 0.6591. However, progressing too far to very hard samples degrades accuracy, indicating a non‑monotonic benefit of difficulty progression.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Jun 5

Level Up: Defining and Exploiting Transitional Problems for Curriculum Learning

arXiv:2603. 13761v2 Announce Type: replace Abstract: Curriculum learning--ordering training examples in a sequence to aid machine learning--takes inspiration from human learning, but has not gained widespread acceptance.

By Amogh Inamdar, Zhenwei Tang, Ashton Anderson, Richard Zemel

arXiv AI

Oct 1

Unlearnable, or Unmeasured? On the Reliability of Difficulty Labels in RLVR

arXiv:2609.40115v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become an important approach for improving reasoning during post-training. Recent work sugges...

By Chandak Chakma, Syed Nazmus Sakib, Nafiul Haque, Shifat E. Arman

reinforcement-learning

arXiv AI

Jul 2

GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity

arXiv:2607. 00152v1 Announce Type: cross Abstract: Three of the most popular methods for training language models to reason look like three different tricks.

By Yong Yi Bay, Kathleen A. Yearick

llms

arXiv AI

Aug 6

Revisiting Generalization Across Difficulty Levels: It's Not So Easy

arXiv:2511. 21692v3 Announce Type: replace-cross Abstract: We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and evaluation.

By Yeganeh Kordi, Nihal V. Nayak, Max Zuo, Ilana Nguyen, Stephen H. Bach

llms

arXiv Machine Learning

Jul 31

Kalman Meets Curriculum: Efficient Dynamic Prompt Selection for Adaptive RL Finetuning

arXiv:2607. 27610v1 Announce Type: new Abstract: Reinforcement learning (RL) finetuning significantly enhances the reasoning capabilities of large language models (LLMs), yet its effectiveness critically depends on selecting prompts of appropriate difficulty for the current policy.

By Haodong Zhu, Yangyang Ren, Yanjing Li, Sheng Xu, Haiguang Liu, Linlin Yang, Baochang Zhang

llms reinforcement-learning fine-tuning benchmarks

arXiv AI

Jun 17

Confusion-Aware Transfer Teacher Curriculum Learning Framework: Disentangling Scoring and Pacing Effects

arXiv:2606. 17706v1 Announce Type: cross Abstract: Curriculum learning couples two design choices, how samples are scored by difficulty and how harder samples are paced into training, making it difficult to attribute observed gains to either component.

By Savini Kommalage, Sanka Mohottala, Asiri Gawesha, Dulara Madhusanka, Menan Velayuthan, Dharshana Kasthurirathna, Mahima Milinda Alwis Weerasinghe, Charith Abhayaratne

One email a morning, machine-written

Related stories

Level Up: Defining and Exploiting Transitional Problems for Curriculum Learning

Unlearnable, or Unmeasured? On the Reliability of Difficulty Labels in RLVR

GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity

Revisiting Generalization Across Difficulty Levels: It's Not So Easy

Kalman Meets Curriculum: Efficient Dynamic Prompt Selection for Adaptive RL Finetuning

Confusion-Aware Transfer Teacher Curriculum Learning Framework: Disentangling Scoring and Pacing Effects