2.8m Gmail.txt -

To break the plateau, the authors implement a two-stage Reinforcement Learning (RL) process [11].

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

) to ensure the generated code matches the visual intent [11]. To break the plateau, the authors implement a

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

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