Differentiable Search Indices (DSIs) encode a corpus of documents in the...
Scaling language models improves performance but comes with significant
...
There have been a lot of interest in the scaling properties of Transform...
Data-to-text generation focuses on generating fluent natural language
re...
There remain many open questions pertaining to the scaling behaviour of
...
Transformers do not scale very well to long sequence lengths largely bec...
Generating fluent natural language responses from structured semantic
re...
Many state-of-the-art neural models for NLP are heavily parameterized an...
This paper tackles the problem of reading comprehension over long narrat...
This paper explores the problem of ranking short social media posts with...
Despite substantial interest in applications of neural networks to
infor...
Most work on natural language question answering today focuses on answer...
Time is an important relevance signal when searching streams of social m...