[Question] Loss computation in movielens examples

Hi, I'm trying to understand the loss computation for the movielens retrieval example. In case of movielens dataset there are ~900 users and ~1600 movies, and when we train the two tower model considering user(U_A)-item(I_A) pairs, we consider only the current U_A-I_A pair as positive (using tf.eye () for labels) and penalizing every other U_A-I_{!A} combinations in the batch. My question is:

1. One user has interacted with multiple items by penalizing all other User-Item pairs in batch, we are also penalizing some U_A-I_B pairs that are positive
2. On increasing the batch size, the contributions from positive pairs reduces significantly

Start with the MovieLens retrieval example and inspect the two-tower loss setup, including the tf.eye labels and in-batch combinations. Done means documenting how multiple user-item interactions and batch size affect positive and negative contributions, with the explanation grounded in the example.