Mixture of Experts layers (MoEs) enable efficient scaling of language models through conditional computation. This paper presents a detailed empirical study of how autoregressive MoE language models scale in comparison with dense models in a wide range of settings: in- and out-of-domain language modeling, zero- and few-shot priming, and full-shot fine-tuning. With the exception of fine-tuning, we find MoEs to be substantially more compute efficient. At more modest training budgets, MoEs can match the performance of dense models using ~4 times less compute. This gap narrows at scale, but our largest MoE model (1.1T parameters) consistently outperforms a compute-equivalent dense model (6.7B parameters). Overall, this performance gap varies greatly across tasks and domains, suggesting that MoE and dense models generalize differently in ways that are worthy of future study. We make our code and models publicly available for research use.

The state of the art on many NLP tasks is currently achieved by large pre-trained language models, which require a considerable amount of computation. We aim to reduce the inference cost in a setting where many different predictions are made on a single piece of text. In that case, computational cost during inference can be amortized over the different predictions (tasks) using a shared text encoder. We compare approaches for training such an encoder and show that encoders pre-trained over multiple tasks generalize well to unseen tasks. We also compare ways of extracting fixed- and limited-size representations from this encoder, including pooling features extracted from multiple layers or positions. Our best approach compares favorably to knowledge distillation, achieving higher accuracy and lower computational cost once the system is handling around 7 tasks. Further, we show that through binary quantization, we can reduce the size of the extracted representations by a factor of 16 to store them for later use. The resulting method offers a compelling solution for using large-scale pre-trained models at a fraction of the computational cost when multiple tasks are performed on the same text.

In this paper, we introduce NeuralClassifier, a toolkit for neural hierarchical multi-label text classification. NeuralClassifier is designed for quick implementation of neural models for hierarchical multi-label classification task, which is more challenging and common in real-world scenarios. A salient feature is that NeuralClassifier currently provides a variety of text encoders, such as FastText, TextCNN, TextRNN, RCNN, VDCNN, DPCNN, DRNN, AttentiveConvNet and Transformer encoder, etc. It also supports other text classification scenarios, including binary-class and multi-class classification. Built on PyTorch, the core operations are calculated in batch, making the toolkit efficient with the acceleration of GPU. Experiments show that models built in our toolkit achieve comparable performance with reported results in the literature.