Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for Reasoning

arXiv:2602.14868 · cs.LG, cs.AI · Submitted 2026-02-16 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-02-16

Updated: 2026-10-02

Comments: 42 pages, 23 figures

Code: https://github.com/huggingface/trl

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• GRPO — 2× in this paper · explained in 6 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• Phi-4 Technical Report
• RL for Reasoning by Adaptively Revealing Rationales
• Self-Evolving Curriculum for LLM Reasoning
• Training Verifiers to Solve Math Word Problems
• DeepSeek-V3 Technical Report
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling
• Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation
• AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset
• Show Your Work: Scratchpads for Intermediate Computation with Language Models
• 2 OLMo 2 Furious
• Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoning
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
• DUMP: Automated Distribution-Level Curriculum Learning for RL-based LLM Post-training
• Reinforce-Ada: An Adaptive Sampling Framework under Non-linear RL Objectives
• Qwen2.5 Technical Report
• Qwen3 Technical Report
• DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks