Search Results for author:
Found 3 papers, 3 papers with code
Memorizing Transformers
Language models typically need to be trained or finetuned in order to acquire new knowledge, which involves updating their weights.
Block-Recurrent Transformers
It is merely a transformer layer: it uses self-attention and cross-attention to efficiently compute a recurrent function over a large set of state vectors and tokens.
Deep Learning with Dynamic Computation Graphs
However, since the computation graph has a different shape and size for every input, such networks do not directly support batched training or inference.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
