The pre-trained language models have achieved great successes in various natural language understanding (NLU) tasks due to its capacity to capture the deep contextualized information in text by pre-training on large-scale corpora. In this technical report, we present our practice of pre-training language models named NEZHA (NEural contextualiZed representation for CHinese lAnguage understanding) on Chinese corpora and finetuning for the Chinese NLU tasks. The current version of NEZHA is based on BERT [1] with a collection of proven improvements, which include Functional Relative Positional Encoding as an effective positional encoding scheme, Whole Word Masking strategy, Mixed Precision Training and the LAMB Optimizer in training the models. The experimental results show that NEZHA achieves the state-of-the-art performances when finetuned on several representative Chinese tasks, including named entity recognition (People's Daily NER), sentence matching (LCQMC), Chinese sentiment classification (ChnSenti) and natural language inference (XNLI).

No abstract

The pre-trained language models have achieved great successes in various natural language understanding (NLU) tasks due to its capacity to capture the deep contextualized information in text by pre-training on large-scale corpora. One of the fundamental components in pre-trained language models is the vocabulary, especially for training multilingual models on many different languages. In the technical report, we present our practices on training multilingual pre-trained language models with BBPE: Byte-Level BPE (i.e., Byte Pair Encoding). BBPE has been adopted by pretrained language models like GPT-2/3 [1, 2] and Roberta [3] and its usage in machine translation has been discussed in [4]. We compared the byte-level vocabulary with the character-level vocabulary adopted in Google's multilingual BERT model through intensive case studies on the tokenization in a variety of languages. In the experiment, we adopted the architecture of NEZHA [5] as the underlying pre-trained language model and the results show that NEZHA trained with byte-level subwords consistently outperforms Google multilingual BERT and vanilla NEZHA by a notable margin in several multilingual NLU tasks. We release the source code of our byte-level vocabulary building tools and the multilingual pre-trained language models at the URLs 12 .

Stacked self-attention models receive widespread attention, due to its ability of capturing global dependency among words. However, the stacking of many layers and components generates huge parameters, leading to low parameter efficiency. In response to this issue, we propose a lightweight architecture named Continuous Self-Attention models with neural ODE networks (CSAODE). In CSAODE, continuous dynamical models (i.e., neural ODEs) are coupled with our proposed self-attention block to form a self-attention ODE solver. This solver continuously calculates and optimizes the hidden states via only one layer of parameters to improve the parameter efficiency. In addition, we design a novel accelerated continuous dynamical model to reduce computing costs, and integrate it in CSAODE. Moreover, since the original self-attention ignores local information, CSAODE makes use of N-gram convolution to encode local representations, and a fusion layer with only two trainable scalars are designed for generating sentence vectors. We perform a series of experiments on text classification, neural language inference (NLI) and text matching tasks. With fewer parameters, CSAODE outperforms state-of-the-art models on text classification tasks (e.g., 1.3% accuracy improved on SUBJ task), and has competitive performances for NLI and text matching tasks as well.