: Increasing data from 2M to 2.8M results in no further performance gains, confirming the plateau [22]. Multimodal Structured Reinforcement Learning (MSRL) :
: The SFT stage requires 60 hours of training on 16 H800 GPUs . The RL stages take an additional 34 hours on 24 H800 GPUs [11]. 2.8M GMAIL.txt
The paper addresses the "SFT plateau," a phenomenon where Supervised Fine-Tuning (SFT) performance on Large Language Models (LLMs) stops improving even as the dataset size increases [11, 22]. The authors use a specific of chart-to-code data to demonstrate this limitation and propose Multimodal Structured Reinforcement Learning (MSRL) as a solution [11, 22]. 2. Methodology Supervised Fine-Tuning (SFT) Phase : Baseline Model : Qwen2.5-VL-7B-Instruct [11, 22]. : Increasing data from 2M to 2
) to ensure the generated code matches the visual intent [11]. The paper addresses the "SFT plateau," a phenomenon
: The model is tested on subsets ranging from 200k to 2.8 million samples.
The paper demonstrates that MSRL significantly outperforms pure SFT models by optimizing for both textual structure and visual fidelity, effectively surpassing the performance limit reached at 2.8M SFT samples [11, 25]. MSRL Stage Max Dataset Size 2.8 million samples [11, 22] 33k curated samples [11] GPU Requirement 16 H800 GPUs [11] 24 H800 GPUs [11] Training Goal Min. Negative Log-Likelihood [22] Hybrid Text-Visual Reward [11] Outcome Performance Plateaus [22] Breaks SFT Performance Limit [11]