Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information

• URL: http://arxiv.org/abs/2607.19313v1
• Date: Tue, 21 Jul 2026 17:28:40 GMT
• Title: Off-Context GRPO: Learning to Reason on Hard Problems using Privileged Information
• Abstract summary: Reinforcement learning with verifiable rewards (RLVR) improves reasoning in large language models.<n>Providing privileged guidance during training, such as solution prefixes, can help overcome this learning cliff.<n>We introduce Off-Context GRPO, a minimally modified variant of GRPO that uses guided rollouts but applies an importance-corrected objective.
• Score: 7.967640756611035
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with verifiable rewards (RLVR) improves reasoning in large language models. Yet, typical RLVR approaches fail on difficult problems: when a model cannot generate any correct solutions, it receives \textit{zero} learning signal. Providing privileged guidance during training, such as solution prefixes, can help overcome this learning cliff by steering the model towards {correct solutions with non-zero reward}. {We call these rollouts \textit{off-context}: they are generated from a training prompt that contains privileged guidance, while the target objective is defined by the original prompt without that guidance.} {We introduce} Off-Context GRPO (OC-GRPO), a minimally modified variant of GRPO that uses guided rollouts but applies an importance-corrected objective to steer the update back toward the original unguided objective, avoiding the mismatch that destabilizes uncorrected guided training. Empirically, our algorithm achieves a 3.9\% absolute improvement (13.8\% relative gain) over vanilla GRPO on average across standard mathematical reasoning benchmarks with negligible additional cost.

Related papers

• TD-Grokking: Learning from Zero-Reward Problems by Training-Time Decomposition [10.882920019894451]
  We propose TD-Grokking, a training-time decomposition framework for zero-reward problems.<n>We show that TD-Grokking outperforms vanilla GRPO as well as all baseline approaches.
  arXiv  Detail & Related papers  (2026-06-03T06:40:04Z)
• Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards [73.44333771806282]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective paradigm for improving the reasoning capabilities of large language models.<n>We propose Correction-Oriented Policy Optimization (CIPO), a simple and effective extension to RLVR that converts on-policy failed trajectories into correction-oriented supervision.<n>CIPO improves learning effectiveness while explicitly enhancing the model's ability to correct its own errors.
  arXiv  Detail & Related papers  (2026-05-14T08:22:21Z)
• Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data [55.84428098924793]
  We propose Constrained Uniform Top-K Sampling (CUTS), a parameter-free decoding strategy enforcing structure-preserving exploration.<n>We integrate this into Mixed-CUTS, a training framework synergizing exploitative and exploratory rollouts to amplify intra-group advantage variance.<n> Notably, Mixed-CUTS improves Pass@1 accuracy on the challenging AIME25 benchmark by up to 15.1% over standard GRPO.
  arXiv  Detail & Related papers  (2026-04-20T16:43:28Z)
• Cog-DRIFT: Exploration on Adaptively Reformulated Instances Enables Learning from Hard Reasoning Problems [70.94268452442147]
  Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of LLMs.<n>We propose a simple yet effective solution based on task reformulation.<n>We introduce Cog-DRIFT, a framework that constructs reformulated variants and organizes them into an adaptive curriculum.
  arXiv  Detail & Related papers  (2026-04-06T15:38:38Z)
• GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning [55.03441672267886]
  We propose GradAlign, a gradient-aligned data selection method for reinforcement learning.<n>We evaluate GradAlign across three data regimes: unreliable reward signals, distribution imbalance, and low-utility training corpus.
  arXiv  Detail & Related papers  (2026-02-25T01:54:50Z)
• Knowing the Answer Isn't Enough: Fixing Reasoning Path Failures in LVLMs [85.37131922131657]
  We reveal a critical yet underexplored flaw in Large Vision-Language Models (LVLMs)<n>Even when these models know the correct answer, they frequently arrive there through incorrect reasoning paths.<n>We propose PSO (Path-Select Optimization), a two-stage post-training framework designed to enhance both the reasoning performance and stability of existing LVLMs.
  arXiv  Detail & Related papers  (2025-12-06T03:02:55Z)
• HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness [49.72591739116668]
  Reinforcement Learning (RL) has become a key driver for enhancing the long chain-of-thought (CoT) reasoning capabilities of Large Language Models (LLMs)<n>However, prevalent methods like GRPO often fail when task difficulty exceeds the model's capacity, leading to reward sparsity and inefficient training.<n>We propose HINT: Helping Ineffective rollouts Navigate Towards effectiveness, an adaptive hinting framework.
  arXiv  Detail & Related papers  (2025-10-10T13:42:03Z)
• Adaptive Guidance Accelerates Reinforcement Learning of Reasoning Models [3.207886496235499]
  We study the process through which reasoning models trained with reinforcement learning on verifiable rewards (RLVR) can learn to solve new problems.<n>We find that RLVR drives performance in two main ways: (1) by compressing pass@$k$ into pass@1 and (2) via "capability gain" in which models learn to solve new problems that they previously could not solve even at high $k$.
  arXiv  Detail & Related papers  (2025-06-16T19:03:06Z)
• Rewarding the Unlikely: Lifting GRPO Beyond Distribution Sharpening [36.81125165911328]
  Reinforcement learning is emerging as a primary driver for improving language model reasoning capabilities.<n>We investigate whether current reinforcement learning algorithms merely sharpen the base model's distribution around problems it can already solve.<n>We show that unlikeliness reward mitigates rank bias and improves pass@$N$ across a large range of $N$ in both synthetic and real theorem proving settings.
  arXiv  Detail & Related papers  (2025-06-03T01:15:15Z)
• GRPO-LEAD: A Difficulty-Aware Reinforcement Learning Approach for Concise Mathematical Reasoning in Language Models [0.3831554157916835]
  Group Relative Policy Optimization ( GRPO) is widely adopted by R1-like reasoning models.<n>We propose GRPO-LEAD, enhancing GRPO with: (1) length-regularized rewards to encourage conciseness while maintaining accuracy; (2) explicit penalties for incorrect solutions to improve model precision; and (3) difficulty-aware advantage reweighting for robust generalization on challenging problems.<n>Our approach achieves state-of-the-art performance for 14B-scale models, underscoring the synergy of our methods with appropriate model scale and high-quality data.
  arXiv  Detail & Related papers  (2025-04-13T19:07:45Z)

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.