2.8m Gmail.txt Site

: 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].

: Uses 11k pairs with a balance of textual and visual rewards ( 2.8M GMAIL.txt

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] : The SFT stage requires 60 hours of

) to ensure the generated code matches the visual intent [11]. MSRL Stage Max Dataset Size 2

: The model is tested on subsets ranging from 200k to 2.8 million samples.

) used in the RL stages or the used to measure the success of the 2.8M dataset?