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).
The formation of dust particles in low-pressure plasmas is a 3-step process. The first one corresponds to nucleation and growth of nanoparticles by chain reactions between ions and gas molecules, the second one is agglomeration of the nanoparticles to form larger particles, and finally, the particles grow by radical deposition on their surfaces. In this work, the nucleation time for carbon dust particles was studied in low pressure acetylene/argon radio frequency (RF) plasmas. Since the self-bias voltage on a powered electrode was drastically affected by the transition from the nucleation to the agglomeration phases, the nucleation time was measured by observing the self-bias voltage time evolution. The nucleation time increases with the gas temperature and decreases when the gas pressure and the RF power are increased. A kinetic model, involving balance between diffusion and charging times of the nanoparticles as well as the chain reactions, is used to explain the exponential dependence of the nucleation time on the gas temperature. The balance between the times was especially indispensable to get good agreement between the model and the experimental results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.