In open-domain question answering, dense passage retrieval has become a new paradigm to retrieve relevant passages for finding answers. Typically, the dual-encoder architecture is adopted to learn dense representations of questions and passages for semantic matching. However, it is difficult to effectively train a dual-encoder due to the challenges including the discrepancy between training and inference, the existence of unlabeled positives and limited training data. To address these challenges, we propose an optimized training approach, called RocketQA, to improving dense passage retrieval. We make three major technical contributions in RocketQA, namely crossbatch negatives, denoised hard negatives and data augmentation. The experiment results show that RocketQA significantly outperforms previous state-of-the-art models on both MS-MARCO and Natural Questions. We also conduct extensive experiments to examine the effectiveness of the three strategies in RocketQA. Besides, we demonstrate that the performance of end-to-end QA can be improved based on our RocketQA retriever 1 .

This paper introduces DuReader, a new large-scale, open-domain Chinese machine reading comprehension (MRC) dataset, designed to address real-world MRC. DuReader has three advantages over previous MRC datasets: (1) data sources: questions and documents are based on Baidu Search and Baidu Zhidao 1 ; answers are manually generated.(2) question types: it provides rich annotations for more question types, especially yes-no and opinion questions, that leaves more opportunity for the research community. (3) scale: it contains 200K questions, 420K answers and 1M documents; it is the largest Chinese MRC dataset so far. Experiments show that human performance is well above current state-of-the-art baseline systems, leaving plenty of room for the community to make improvements. To help the community make these improvements, both DuReader 2 and baseline systems 3 have been posted online. We also organize a shared competition to encourage the exploration of more models. Since the release of the task, there are significant improvements over the baselines.

Current evaluation metrics to question answering based machine reading comprehension (MRC) systems generally focus on the lexical overlap between candidate and reference answers, such as ROUGE and BLEU. However, bias may appear when these metrics are used for specific question types, especially questions inquiring yes-no opinions and entity lists. In this paper, we make adaptations on the metrics to better correlate n-gram overlap with the human judgment for answers to these two question types. Statistical analysis proves the effectiveness of our approach. Our adaptations may provide positive guidance for the development of realscene MRC systems.

The degradation model and the remaining useful life (RUL) prediction are two key metrics for machine health prognostics. However, the precision of RUL prediction highly depends on the validity of the degradation model. Consequently, the uncertainty of RUL prediction based on the two-phase degradation process may cause the estimated product reliability to differ from its actual value and lead to a faulty predictive maintenance strategy. However, recent studies have rarely compared the RUL predictions based on two-phase and single-phase degradation models. In this study, a real-time RUL prediction approach was developed for a two-phase linear Wiener degradation process. A two-phase Wiener degradation model was proposed along with an estimation method for the change point and model parameters. Based on a state-space model and Markov chain Monte Carlo, the drift coefficient of the degradation model was updated, and a real-time RUL prediction model was developed. To examine whether the twophase Wiener degradation model was incorrectly assumed to be single-phase, a case study was conducted to analyze RUL prediction based on the original and incorrect models. The results demonstrated that to obtain a more accurate RUL prediction, a sufficient number of samples should be used in addition to updating the drift coefficient of the degradation model when new measurement data are available.