During real-time production in industrial Internet of Things systems, equipment changes its operating speed due to changing operating conditions. And dynamic speed changes of rotating machinery under fluctuating workloads often lead to domain changes of vibration signals, which will directly lead to degradation of fault diagnostic model performance. Furthermore, the acquisition of data from multiple domains in real industrial scenarios is challenging due to the expense of collecting data from all possible working conditions. Consequently, applying diagnostic models trained using a single-source domain directly to an unknown target domain is a very challenging single domain generalization problem. Therefore, a generic single-source domain generalization framework via wavelet packet augmentation and pseudo-domain generation for fault diagnosis under unknown operating conditions is proposed in this paper. Pseudo-domain generation involves augmenting single-source domain by integrating data generetion model, thereby enhancing prediction accuracy. Furthermore, a wavelet packet augmentation method is proposed. Initially, the original signal is decomposed to obtain high and low frequency information. Subsequently, the high and low frequency information within the batch are linearly interpolated, respectively. Consequently, the interpolated high and low frequency information is then reconstructed to yield enhanced samples. The experimental results on four datasets show that the proposed framework can effectively improve the robustness of the generalization ability of fault diagnosis under unknown operating environments.

Since the development of artificial intelligence technology, E-Commerce has gradually become one of the world's largest commercial markets. Within this domain, sales events, which are based on sociological mechanisms, play a significant role. E-Commerce platforms frequently offer sales and promotions to encourage users to purchase items, leading to significant changes in live environments. Learning-To-Rank (LTR) is a crucial component of E-Commerce search and recommendations, and substantial efforts have been devoted to this area. However, existing methods often assume an independent and identically distributed data setting, which does not account for the evolving distribution of online systems beyond online finetuning strategies. This limitation can lead to inaccurate predictions of user behaviors during sales events, resulting in significant loss of revenue. In addition, models must readjust themselves once sales have concluded in order to eliminate any effects caused by the sales events, leading to further regret. To address these limitations, we introduce a long-term E-Commerce search data set specifically designed to incubate LTR algorithms during such sales events, with the objective of advancing the capabilities of E-Commerce search engines. Our investigation focuses on typical industry practices and aims to identify potential solutions to address these challenges.

Abstract Turbo codes play a crucial role in wireless communication systems, and their compiled code structures are key factors affecting the performance of the entire communication system. As a result, the study of turbo compiled code structures has been a focal point for researchers. The iterative decoding of turbo code structures has multiple limitations and large storage resource consumption, leading to poor system anti‐interference ability and a rapid increase in BER. To address these issues, this paper proposes the mimic turbo compiled code structure (MTCCS) for wireless communication systems. MTCCS is based on the DHR idea, incorporating dynamic, heterogeneous, and redundancy characteristics. Dynamicity is achieved through a dynamic scheduling algorithm based on abnormal feedback information. Heterogeneity is achieved through a codec component collection design method based on intrinsic and extrinsic heterogeneity. Redundancy is achieved through a majority voting algorithm. At the beginning of information transmission, MTCCS randomly selects heterogeneous codecs from the heterogeneous codec collection to enter the runtime pool. After the information transmission is complete, the majority voting algorithm is used to adjudicate the multi‐mode output of the codecs, resulting in a relatively accurate decoding outcome. Meanwhile, the dynamic scheduling module calculates the abnormal feedback information of each codec and accordingly dynamically schedules the mimic turbo codecs to replace the abnormal ones. Through the above process, MTCCS realizes the adaptive compilation code and improves the anti‐interference ability of turbo code. Simulation experiments are conducted on MTCCS in both non‐interference and interference scenarios. Simulation experiments show that MTCCS introducing the DHR idea achieves a balance between anti‐interference and decoding performance. It effectively addresses the issue of poor anti‐interference ability in turbo codes, and the decoding performance of MTCCS is superior to that of the previous single conventional turbo codes.

Recent machine reading comprehension datasets such as ReClor and LogiQA require performing logical reasoning over text.Conventional neural models are insufficient for logical reasoning, while symbolic reasoners cannot directly apply to text.To meet the challenge, we present a neural-symbolic approach which, to predict an answer, passes messages over a graph representing logical relations between text units.It incorporates an adaptive logic graph network (AdaLoGN) which adaptively infers logical relations to extend the graph and, essentially, realizes mutual and iterative reinforcement between neural and symbolic reasoning.We also implement a novel subgraph-to-node message passing mechanism to enhance context-option interaction for answering multiple-choice questions.Our approach shows promising results on ReClor and LogiQA.

Scenario-based question answering (SQA) has attracted an increasing research interest. Compared with the well-studied machine reading comprehension (MRC), SQA is a more challenging task: a scenario may contain not only a textual passage to read but also structured data like tables, i.e., tabular scenario based question answering (TSQA). AI applications of TSQA such as answering multiple-choice questions in high-school exams require synthesizing data in multiple cells and combining tables with texts and domain knowledge to infer answers. To support the study of this task, we construct GeoTSQA. This dataset contains 1k real questions contextualized by tabular scenarios in the geography domain. To solve the task, we extend state-of-the-art MRC methods with TTGen, a novel table-to-text generator. It generates sentences from variously synthesized tabular data and feeds the downstream MRC method with the most useful sentences. Its sentence ranking model fuses the information in the scenario, question, and domain knowledge. Our approach outperforms a variety of strong baseline methods on GeoTSQA.

Semantic parsing transforms a natural language question into a formal query over a knowledge base. Many existing methods rely on syntactic parsing like dependencies. However, the accuracy of producing such expressive formalisms is not satisfying on long complex questions. In this paper, we propose a novel skeleton grammar to represent the high-level structure of a complex question. This dedicated coarse-grained formalism with a BERT-based parsing algorithm helps to improve the accuracy of the downstream fine-grained semantic parsing. Besides, to align the structure of a question with the structure of a knowledge base, our multi-strategy method combines sentence-level and word-level semantics. Our approach shows promising performance on several datasets.

Although neural table-to-text models have achieved remarkable progress with the help of largescale datasets, they suffer insufficient learning problem with limited training data.Recently, pretrained language models show potential in few-shot learning with linguistic knowledge learnt from pretraining on large-scale corpus.However, benefiting table-to-text generation in few-shot setting with the powerful pretrained language model faces three challenges, including (1) the gap between the task's structured input and the natural language input for pretraining language model.(2) The lack of modeling for table structure and ( 3) improving text fidelity with less incorrect expressions that are contradicting to the table.To address aforementioned problems, we propose TableGPT for table-to-text generation.At first, we utilize table transformation module with template to rewrite structured table in natural language as input for GPT-2.In addition, we exploit multi-task learning with two auxiliary tasks that preserve table's structural information by reconstructing the structure from GPT-2's representation and improving the text's fidelity with content matching task aligning the table and information in the generated text.By experimenting on Humans, Songs and Books, three few-shot table-to-text datasets in different domains, our model outperforms existing systems on most few-shot settings.

Complex questions in reading comprehension tasks require integrating information from multiple sentences. In this work, to answer such questions involving temporal and causal relations, we generate event graphs from text based on dependencies, and rank answers by aligning event graphs. In particular, the alignments are constrained by graph-based reasoning to ensure temporal and causal agreement. Our focused approach self-adaptively complements existing solutions; it is automatically triggered only when applicable. Experiments on RACE and MCTest show that state-of-the-art methods are notably improved by using our approach as an add-on.