Core Viewpoint - The article discusses the relationship between mathematical reasoning capabilities of large language models (LLMs) and their ability to transfer these skills to other tasks, highlighting that models trained with reinforcement learning (RL) show better transferability compared to those trained with supervised fine-tuning (SFT) [4][11]. Group 1: Mathematical Reasoning and Transferability - Research indicates that only models trained with RL can effectively transfer mathematical reasoning skills to other tasks, while SFT models show limited or no transfer [4][11]. - A Transferability Index (TI) is introduced to quantify the extent to which improvements in mathematical reasoning can be applied to other reasoning and non-reasoning tasks [8][9]. - If TI is greater than 0, it indicates a positive transfer effect to other tasks; if less than 0, it indicates negative transfer [9]. Group 2: Experimental Findings - The study evaluated over 20 models across various tasks, including mathematical reasoning, other reasoning tasks (like medical reasoning), and non-reasoning tasks (like common-sense dialogue) [7]. - Results show that models fine-tuned with RL consistently achieve higher transferability metrics across reasoning and non-reasoning tasks, while SFT models often experience negative transfer in non-reasoning tasks [11]. Group 3: Model Representation and Performance - PCA analysis reveals that RL fine-tuned models exhibit minimal shifts in representation space, indicating they retain previously learned knowledge while enhancing performance in specific domains [15]. - RL models demonstrate lower KL divergence in reasoning and non-reasoning tasks compared to SFT models, suggesting more stable and precise representation updates [16][18]. - The findings suggest that RL is crucial for achieving transferable reasoning capabilities in LLMs, marking another victory for reinforcement learning in this context [19].

Core Viewpoint - The article introduces the Supervised Reinforcement Fine-Tuning (SRFT) method, which combines supervised fine-tuning (SFT) and reinforcement learning (RL) in a single-stage approach to enhance the reasoning performance of large language models (LLMs) [1][22]. Group 1: Methodology - SRFT employs a dual strategy design to effectively utilize demonstration data, incorporating both SFT for coarse-grained behavior policy approximation and RL for fine-grained policy refinement [23][24]. - The method introduces an entropy-aware adaptive weighting mechanism to balance the influence of SFT and RL, ensuring stable training dynamics [29][44]. - SRFT achieves a significant improvement in training efficiency, speeding up the process by 2.28 times compared to traditional sequential methods [21][44]. Group 2: Performance Results - SRFT demonstrates an average accuracy of 59.1% across five mathematical reasoning tasks, outperforming the zero-RL baseline by 9.0% [4][47]. - In out-of-distribution tasks, SRFT achieves an average accuracy of 62.5%, surpassing the best baseline by 10.9% [4][47]. - The method shows superior generalization capabilities, with consistent performance improvements across various benchmarks [47][48]. Group 3: Training Dynamics - The training dynamics of SRFT reveal a more stable and efficient learning process, with a gradual increase in response length indicating a deeper reasoning process [48]. - SRFT maintains a more stable entropy during training, allowing for continued exploration, unlike pure RL which exhibits rapid entropy decline [20][48]. - The analysis of training trajectories indicates that SRFT effectively balances knowledge acquisition and self-exploration without excessive deviation from the initial model [15][45].

Core Insights - The article discusses the limitations of the "Supervised Fine-Tuning (SFT) + Reinforcement Learning (RL)" paradigm in developing large vision-language models (LVLM), suggesting that SFT may hinder learning and lead to superficial reasoning paths, while RL promotes genuine multimodal reasoning [3][11][21]. Group 1: Research Findings - A study from the University of California, Santa Cruz, and the University of Texas at Dallas reveals that SFT can obstruct learning, often resulting in "pseudo-reasoning paths" that lack depth [3][11]. - The research team created the VLAA-Thinking dataset to systematically investigate the roles of SFT and RL in multimodal reasoning, highlighting the unique contributions of each method [4][8]. - The findings indicate that while SFT improves performance on standard tasks, it falls short in enhancing complex reasoning capabilities, leading to a 47% relative performance decline in a 7B model [11][13]. Group 2: Data and Methodology - The VLAA-Thinking dataset comprises 203,182 samples, with 126,413 for SFT and 25,195 for RL, designed to facilitate high-quality reasoning chains [5][6]. - The research employed a six-stage data processing workflow to effectively transfer reasoning capabilities from pure text models to LVLMs [6][8]. - A mixed reward function was innovatively designed within the GRPO framework to optimize RL in visual contexts, incorporating various reward types for different problem categories [8][19]. Group 3: Performance Analysis - The study found that SFT's imitative reasoning patterns can limit the exploration space during the RL phase, suggesting that direct learning from reward signals is more effective [15][26]. - Models trained solely with GRPO outperformed those that underwent SFT, with the VLAA-Thinker-Qwen2.5-VL-3B model ranking first in the Open LMM reasoning leaderboard for 4B models, achieving a 1.8% record improvement [15][31]. - The analysis revealed that response length and reward scores do not correlate significantly with performance, challenging previous assumptions about their relationship [24][26]. Group 4: Implications for Future Research - The findings suggest that SFT is currently incompatible with GRPO in the context of multimodal reasoning, potentially damaging the performance of both foundational and instruction-tuned LVLMs [21][22]. - The research emphasizes the need for high-quality instruction tuning to enhance model performance in RL settings, indicating that better instruction tuning leads to improved reasoning capabilities post-RL training [31].

Core Viewpoint - The article discusses the introduction of a new technology called OThink-MR1, developed by researchers from OPPO Research Institute and Hong Kong University of Science and Technology, which enhances multimodal language models' generalized reasoning capabilities through dynamic reinforcement learning [1][2][29]. Group 1: Technology Overview - OThink-MR1 extends reinforcement learning to multimodal language models, enabling them to better handle complex tasks and new scenarios [1][2]. - The technology addresses the limitations of existing multimodal models that primarily rely on supervised fine-tuning (SFT), which hinders the development of general reasoning abilities [4][5]. - OThink-MR1 employs two core components: dynamic KL divergence strategy (GRPO-D) and a carefully designed reward model, significantly improving learning efficiency and reasoning capabilities [8]. Group 2: Dynamic KL Divergence Strategy - The dynamic KL divergence strategy balances exploration of new strategies and utilization of existing experiences, adapting as training progresses [10][11]. - This approach prevents the model from getting stuck in local optima by encouraging exploration in the early stages and gradually shifting towards leveraging accumulated knowledge [12]. Group 3: Reward Model - The reward model in OThink-MR1 provides two types of rewards: validation accuracy reward and format reward, guiding the model's learning process [13][14]. - These rewards help the model understand its strengths and areas for improvement, promoting targeted learning [15]. Group 4: Experimental Validation - The first experiment demonstrated that incorporating format rewards significantly improved model performance in geometric reasoning tasks, highlighting the importance of both content and format in evaluations [17]. - The second experiment tested the model's cross-task evaluation, showing that the GRPO-D trained model excelled in diverse tasks, unlike models trained with SFT [21][23]. - The third experiment revealed that OThink-MR1's GRPO-D outperformed traditional SFT methods in same-task evaluations, indicating its effectiveness in enhancing model capabilities [28]. Group 5: Future Implications - OThink-MR1 represents a significant advancement in the development of multimodal language models, showcasing the potential of dynamic reinforcement learning to enhance reasoning and generalization abilities [29].