# Decouple to Generalize: Context-First Self-Evolving Learning for Data-Scarce Vision-Language Reasoning arXiv: https://arxiv.org/abs/2512.06835 ยท HTML: https://mlsanity.com/paper/2512.06835/ ## Key facts - Cited by 2 papers in the map (data through 14 Oct 2026). - Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. - Builds on 30 of its 46 references that are in the map. - First posted on arXiv 7 Dec 2025. ## Placed in - LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/) - Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/) - Synthetic data and self-improvement for LLMs / [Self-improving MLLMs and synthetic data](https://mlsanity.com/synth/self-improving-mllms-and-synthetic-data/) ## Builds on (most cited first, 30 of 30) - [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017) - [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024) - [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025) - [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025) - [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025) - [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025) - [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025) - [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024) - [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023) - [Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models](https://mlsanity.com/paper/2503.06749/) (2025) - [VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model](https://mlsanity.com/paper/2504.07615/) (2025) - [Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning](https://mlsanity.com/paper/2105.04165/) (2021) - [VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning](https://mlsanity.com/paper/2504.08837/) (2025) - [R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization](https://mlsanity.com/paper/2503.12937/) (2025) - [MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts](https://mlsanity.com/paper/2404.15159/) (2024) - [Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models](https://mlsanity.com/paper/2503.13939/) (2025) - [LlaSMol: Advancing Large Language Models for Chemistry with a Large-Scale, Comprehensive, High-Quality Instruction Tuning Dataset](https://mlsanity.com/paper/2402.09391/) (2024) - [OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles](https://mlsanity.com/paper/2503.17352/) (2025) - [Improving Vision-Language-Action Model with Online Reinforcement Learning](https://mlsanity.com/paper/2501.16664/) (2025) - [Q-Insight: Understanding Image Quality via Visual Reinforcement Learning](https://mlsanity.com/paper/2503.22679/) (2025) - [Self-Rewarding Vision-Language Model via Reasoning Decomposition](https://mlsanity.com/paper/2508.19652/) (2025) - [Visionary-R1: Mitigating Shortcuts in Visual Reasoning with Reinforcement Learning](https://mlsanity.com/paper/2505.14677/) (2025) - [More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models](https://mlsanity.com/paper/2505.21523/) (2025) - [SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from Experience](https://mlsanity.com/paper/2508.04700/) (2025) - [Beyond Pass@1: Self-Play with Variational Problem Synthesis Sustains RLVR](https://mlsanity.com/paper/2508.14029/) (2025) - [Intern-S1: A Scientific Multimodal Foundation Model](https://mlsanity.com/paper/2508.15763/) (2025) - [CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning](https://mlsanity.com/paper/2509.22647/) (2025) - [UniRL: Self-Improving Unified Multimodal Models via Supervised and Reinforcement Learning](https://mlsanity.com/paper/2505.23380/) (2025) - [ARES: Alternating Reinforcement Learning and Supervised Fine-Tuning for Enhanced Multi-Modal Chain-of-Thought Reasoning Through Diverse AI Feedback](https://mlsanity.com/paper/2407.00087/) (2024) - [Generalizable Geometric Image Caption Synthesis](https://mlsanity.com/paper/2509.15217/) (2025) ## Built on by (most cited first, 2 of 2) - [Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges](https://mlsanity.com/paper/2604.13602/) (2026) - [Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models](https://mlsanity.com/paper/2606.27373/) (2026) ## Abstract (arXiv) Recent vision-language models (VLMs) achieve remarkable reasoning through reinforcement learning (RL), which provides a feasible solution for realizing continuous self-evolving large vision-language models (LVLMs) in the era of experience. However, RL for VLMs requires abundant high-quality multimodal data, especially challenging in specialized domains like chemistry, earth sciences, and multimodal mathematics. Existing strategies such as synthetic data and self-rewarding mechanisms suffer from limited distributions and alignment difficulties, ultimately causing reward hacking: models exploit high-reward patterns, collapsing policy entropy and destabilizing training. We propose DoGe (Decouple to Generalize), a dual-decoupling framework that guides models to first learn from context rather than problem solving by refocusing on the problem context scenarios overlooked by synthetic data methods. By decoupling learning process into dual components (Thinker and Solver), we reasonably quantify the reward signals of this process and propose a two-stage RL post-training approach from freely exploring context to practically solving tasks. Second, to increase the diversity of training data, DoGe constructs an evolving curriculum learning pipeline: an expanded native domain knowledge corpus and an iteratively evolving seed problems pool. Experiments show that our method consistently outperforms the baseline across various benchmarks, providing a scalable pathway for realizing self-evolving LVLMs.