{"slug": "cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and", "title": "CVPO: Enhancing LLM Reinforcement Learning Reasoning via Value-Variance Adaptation and Dynamic Curriculum Learning", "summary": "Researchers propose CVPO (Curriculum-guided Value-Variance Policy Optimization), a new reinforcement learning method that enhances LLM reasoning by adapting to token-level value-variance and question difficulty. The method outperforms the strong baseline VAPO across various math tasks, achieving better performance and stronger exploration. The work is detailed in arXiv:2608.03068v1.", "body_md": "arXiv:2608.03068v1 Announce Type: new\nAbstract: Reinforcement learning (RL) has emerged as an effective method for enhancing the reasoning capabilities of large language models (LLMs). However, existing methods suffer from insufficient precision in feedback on generated answer trajectories and exhibit the phenomenon of problem difficulty drift. To address these challenges, we propose CVPO - Curriculum-guided Value-Variance Policy Optimization. At the response trajectory level, we find that token-level value-variance correlates with exploration intensity. Our theoretical analysis shows this variance bounds policy update magnitude. We then use the estimated trajectory value-variance to quantify the intrinsic randomness in generation. Based on this, we design a variance-aware advantage adjustment mechanism for different reward types. At the question level, we introduce a dynamic curriculum weighting method that adapts to question difficulty. This helps the model focus on tasks matched to its current ability during each training stage. Experimental results show our method outperforms strong value-based baselines like VAPO. It achieves better performance and stronger exploration, enabling more accurate and robust reasoning in language models across various math tasks.", "url": "https://wpnews.pro/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and", "canonical_source": "https://www.machinebrief.com/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-valu-7bus", "published_at": "2026-08-05 04:00:00+00:00", "updated_at": "2026-08-05 05:03:39.906791+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["CVPO", "VAPO", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and", "markdown": "https://wpnews.pro/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and.md", "text": "https://wpnews.pro/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and.txt", "jsonld": "https://wpnews.pro/news/cvpo-enhancing-llm-reinforcement-learning-reasoning-via-value-variance-and.jsonld"}}