相关论文: Optimal Subarchitecture Extraction For BERT

Recurrent neural network (RNN)'s architecture is a key factor influencing its performance. We propose algorithms to optimize hidden sizes under running time constraint. We convert the discrete optimization into a subset selection problem.…

Pre-trained language models like BERT have achieved great success in a wide variety of NLP tasks, while the superior performance comes with high demand in computational resources, which hinders the application in low-latency IR systems. We…

Newly-introduced deep learning architectures, namely BERT, XLNet, RoBERTa and ALBERT, have been proved to be robust on several NLP tasks. However, the datasets trained on these architectures are fixed in terms of size and generalizability.…

This paper describes a memory-efficient transformer model designed to drive a reduction in memory usage and execution time by substantial orders of magnitude without impairing the model's performance near that of the original model.…

BERT-based architectures currently give state-of-the-art performance on many NLP tasks, but little is known about the exact mechanisms that contribute to its success. In the current work, we focus on the interpretation of self-attention,…

Language model pre-training, such as BERT, has achieved remarkable results in many NLP tasks. However, it is unclear why the pre-training-then-fine-tuning paradigm can improve performance and generalization capability across different…

We introduce the CUDA Tensor Transpose (cuTT) library that implements high-performance tensor transposes for NVIDIA GPUs with Kepler and above architectures. cuTT achieves high performance by (a) utilizing two GPU-optimized transpose…

Due to high accuracy, BERT-like models have been widely adopted by text mining and web searching. However, large BERT-like models suffer from inefficient online inference, facing the following two problems on GPUs: (1) their high accuracy…

In recent years, researchers tend to pre-train ever-larger language models to explore the upper limit of deep models. However, large language model pre-training costs intensive computational resources and most of the models are trained from…

Transformer-based language models such as BERT have achieved the state-of-the-art performance on various NLP tasks, but are computationally prohibitive. A recent line of works use various heuristics to successively shorten sequence length…

Transformer architectures based on the attention mechanism have revolutionized natural language processing (NLP), driving major breakthroughs across virtually every NLP task. However, their substantial memory and computational requirements…

As language models have grown in parameters and layers, it has become much harder to train and infer with them on single GPUs. This is severely restricting the availability of large language models such as GPT-3, BERT-Large, and many…

Large-scale pre-trained language models have shown remarkable results in diverse NLP applications. Unfortunately, these performance gains have been accompanied by a significant increase in computation time and model size, stressing the need…

We present a new semi-external algorithm that builds the Burrows--Wheeler transform variant of Bauer et al. (a.k.a., BCR BWT) in linear expected time. Our method uses compression techniques to reduce computational costs when the input is…

Bayesian Additive Regression Trees (BART) is a nonparametric Bayesian regression technique based on an ensemble of decision trees. It is part of the toolbox of many statisticians. The overall statistical quality of the regression is…

Pre-trained language models have shown remarkable results on various NLP tasks. Nevertheless, due to their bulky size and slow inference speed, it is hard to deploy them on edge devices. In this paper, we have a critical insight that…

Hidden-unit BERT (HuBERT) is a widely-used self-supervised learning (SSL) model in speech processing. However, we argue that its fixed 20ms resolution for hidden representations would not be optimal for various speech-processing tasks since…

Recent progress in the Natural Language Processing domain has given us several State-of-the-Art (SOTA) pretrained models which can be finetuned for specific tasks. These large models with billions of parameters trained on numerous GPUs/TPUs…

Fine-tuning and inference with large Language Models (LM) are generally known to be expensive. Parameter-efficient fine-tuning over pretrained LMs reduces training memory by updating a small number of LM parameters but does not improve…

Although BERT based relation classification (RC) models have achieved significant improvements over the traditional deep learning models, it seems that no consensus can be reached on what is the optimal architecture. Firstly, there are…