Cover Image

Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Parashar, Shubham, Gui, Shurui, Li, Xiner, Ling, Hongyi, Vemuri, Sushil, Olson, Blake, Li, Eric, Zhang, Yu, Caverlee, James, Kalathil, Dileep, Ji, Shuiwang
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Inference-Time Computations for LLM Reasoning and Planning: A Benchmark and Insights
by: Parashar, Shubham, et al.
Published: (2025)

Active Test-Time Adaptation: Theoretical Analyses and An Algorithm
by: Gui, Shurui, et al.
Published: (2024)

Discovering Physics Laws of Dynamical Systems via Invariant Function Learning
by: Gui, Shurui, et al.
Published: (2025)

A Hierarchical Language Model For Interpretable Graph Reasoning
by: Khurana, Sambhav, et al.
Published: (2024)

Adaptive Outer-Loop Control of Quadrotors via Reinforcement Learning
by: Saj, Vishnu, et al.
Published: (2026)

Graph Structure and Feature Extrapolation for Out-of-Distribution Generalization
by: Li, Xiner, et al.
Published: (2023)

Complex LLM Planning via Automated Heuristics Discovery
by: Ling, Hongyi, et al.
Published: (2025)

Fragment and Geometry Aware Tokenization of Molecules for Structure-Based Drug Design Using Language Models
by: Fu, Cong, et al.
Published: (2024)

Mitigating Spurious Correlations in LLMs via Causality-Aware Post-Training
by: Gui, Shurui, et al.
Published: (2025)

Robust LLM Alignment via Distributionally Robust Direct Preference Optimization
by: Xu, Zaiyan, et al.
Published: (2025)

Geometry Informed Tokenization of Molecules for Language Model Generation
by: Li, Xiner, et al.
Published: (2024)

Equivariance via Minimal Frame Averaging for More Symmetries and Efficiency
by: Lin, Yuchao, et al.
Published: (2024)

Federated Ensemble-Directed Offline Reinforcement Learning
by: Rengarajan, Desik, et al.
Published: (2023)

FlowX: Towards Explainable Graph Neural Networks via Message Flows
by: Gui, Shurui, et al.
Published: (2022)

AttackGNN: Red-Teaming GNNs in Hardware Security Using Reinforcement Learning
by: Gohil, Vasudev, et al.
Published: (2024)

Object-Aware 4D Human Motion Generation
by: Gui, Shurui, et al.
Published: (2025)

Rethinking Easy-to-Hard: Limits of Curriculum Learning in Post-Training for Deductive Reasoning
by: Mordig, Maximilian, et al.
Published: (2026)

Invariant Tokenization of Crystalline Materials for Language Model Enabled Generation
by: Yan, Keqiang, et al.
Published: (2025)

Language Models for Controllable DNA Sequence Design
by: Su, Xingyu, et al.
Published: (2025)

PowerMamba: A Deep State Space Model and Comprehensive Benchmark for Time Series Prediction in Electric Power Systems
by: Menati, Ali, et al.
Published: (2024)

MAVIS: Multi-Objective Alignment via Inference-Time Value-Guided Selection
by: Carleton, Jeremy, et al.
Published: (2025)

Risk-Averse Finetuning of Large Language Models
by: Chaudhary, Sapana, et al.
Published: (2025)

Selling Demand Response Using Options
by: Muthirayan, Deepan, et al.
Published: (2019)

Structured Reinforcement Learning for Media Streaming at the Wireless Edge
by: Bura, Archana, et al.
Published: (2024)

The Neglected Tails in Vision-Language Models
by: Parashar, Shubham, et al.
Published: (2024)

From Easy to Hard++: Promoting Differentially Private Image Synthesis Through Spatial-Frequency Curriculum
by: Gong, Chen, et al.
Published: (2026)

Optimistic World Models: Efficient Exploration in Model-Based Deep Reinforcement Learning
by: Mete, Akshay, et al.
Published: (2026)

Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penalty
by: Ling, Zehui, et al.
Published: (2025)

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks
by: Hase, Peter, et al.
Published: (2024)

Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks
by: Gambardella, Andrew, et al.
Published: (2024)

Can an Easy-to-Hard Curriculum Make Reasoning Emerge in Small Language Models? Evidence from a Four-Stage Curriculum on GPT-2
by: Fu, Xiang
Published: (2025)

Reinforcement Learning for Diffusion LLMs with Entropy-Guided Step Selection and Stepwise Advantages
by: Kunde, Vishnu Teja, et al.
Published: (2026)

Abstractions-of-Thought: Intermediate Representations for LLM Reasoning in Hardware Design
by: DeLorenzo, Matthew, et al.
Published: (2025)

Charge Puddles Driven Complex Crossover of Magnetoresistance in Non‐Topological Sulfur Doped Antimony Selenide Nanowires
by: Sushil Kumar, et al.
Published: (2024)

Reward-Guided Iterative Refinement in Diffusion Models at Test-Time with Applications to Protein and DNA Design
by: Uehara, Masatoshi, et al.
Published: (2025)

Dynamic Search for Inference-Time Alignment in Diffusion Models
by: Li, Xiner, et al.
Published: (2025)

Diffusion Blend: Inference-Time Multi-Preference Alignment for Diffusion Models
by: Cheng, Min, et al.
Published: (2025)

Improving Medical Reasoning with Curriculum-Aware Reinforcement Learning
by: Rui, Shaohao, et al.
Published: (2025)

Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges
by: Lee, Nayoung, et al.
Published: (2025)

A Task-Centric Theory for Iterative Self-Improvement with Easy-to-Hard Curricula
by: Liu, Chenruo, et al.
Published: (2026)