Pretraining is the preliminary and fundamental step in developing capabl...
Pretrained multilingual large language models have typically used heuris...
We study the design decisions of publicly available instruction tuning
m...
We present SingSong, a system that generates instrumental music to accom...
We introduce MusicLM, a model generating high-fidelity music from text
d...
Current image generation models struggle to reliably produce well-formed...
While deep learning models have replaced hand-designed features across m...
The internet contains a wealth of knowledge – from the birthdays of
hist...
Multitask prompted finetuning (MTF) has been shown to help large languag...
An ideal music synthesizer should be both interactive and expressive,
ge...
Large pretrained Transformer language models have been shown to exhibit
...
Large language models have been shown to achieve remarkable performance
...
Recent neural network-based language models have benefited greatly from
...
Most widely-used pre-trained language models operate on sequences of tok...
The research community has proposed copious modifications to the Transfo...
We review the EfficientQA competition from NeurIPS 2020. The competition...
It has become common to publish large (billion parameter) language model...
The recent "Text-to-Text Transfer Transformer" (T5) leveraged a unified
...
Neural networks have recently achieved human-level performance on variou...
It has recently been observed that neural language models trained on
uns...
Most generative models of audio directly generate samples in one of two
...
Transfer learning, where a model is first pre-trained on a data-rich tas...
To make music composition more approachable, we designed the first AI-po...
We explore models for translating abstract musical ideas (scores, rhythm...
Machine learning models of music typically break up the task of composit...
Efficient audio synthesis is an inherently difficult machine learning ta...
Generating musical audio directly with neural networks is notoriously
di...
Discovering and exploring the underlying structure of multi-instrumental...
The Variational Autoencoder (VAE) has proven to be an effective model fo...
Deep generative neural networks have proven effective at both conditiona...
We consider the problem of transcribing polyphonic piano music with an
e...
Generative models in vision have seen rapid progress due to algorithmic
...