Topic taxonomy discovery aims at uncovering topics of different abstraction levels and constructing hierarchical relations between them. Unfortunately, most prior work can hardly model semantic scopes of words and topics by holding the Euclidean embedding space assumption. What’s worse, they infer asymmetric hierarchical relations by symmetric distances between topic embeddings. As a result, existing methods suffer from problems of low-quality topics at high abstraction levels and inaccurate hierarchical relations. To alleviate these problems, this paper develops a Box embedding-based Topic Model (BoxTM) that maps words and topics into the box embedding space, where the asymmetric metric is defined to properly infer hierarchical relations among topics. Additionally, our BoxTM explicitly infers upper-level topics based on correlation between specific topics through recursive clustering on topic boxes. Finally, extensive experiments validate high-quality of the topic taxonomy learned by BoxTM.

Word sense disambiguation (WSD) can be viewed as two subtasks: textual word sense disambiguation (Textual-WSD) and visual word sense disambiguation (Visual-WSD). They aim to identify the most semantically relevant senses or images to a given context containing ambiguous target words. However, existing WSD models seldom address these two subtasks jointly due to lack of images in Textual-WSD datasets or lack of senses in Visual-WSD datasets. To bridge this gap, we propose PolCLIP, a unified image-text WSD model. By employing an image-text complementarity strategy, it not only simulates stable diffusion models to generate implicit visual representations for word senses but also simulates image captioning models to provide implicit textual representations for images. Additionally, a disambiguation-oriented image-sense dataset is constructed for the training objective of learning multimodal polysemy representations. To the best of our knowledge, PolCLIP is the first model that can cope with both Textual-WSD and Visual-WSD. Extensive experimental results on benchmarks demonstrate the effectiveness of our method, achieving a 2.53% F1-score increase over the state-of-the-art models on Textual-WSD and a 2.22% HR@1 improvement on Visual-WSD.

Self-attention-based models have achieved remarkable progress in short-text mining. However, the quadratic computational complexities restrict their application in long text processing. Prior works have adopted the chunking strategy to divide long documents into chunks and stack a self-attention backbone with the recurrent structure to extract semantic representation. Such an approach disables parallelization of the attention mechanism, significantly increasing the training cost and raising hardware requirements. Revisiting the self-attention mechanism and the recurrent structure, this paper proposes a novel long-document encoding model, Recurrent Attention Network (RAN), to enable the recurrent operation of self-attention. Combining the advantages from both sides, the well-designed RAN is capable of extracting global semantics in both token-level and document-level representations, making it inherently compatible with both sequential and classification tasks, respectively. Furthermore, RAN is computationally scalable as it supports parallelization on long document processing. Extensive experiments demonstrate the long-text encoding ability of the proposed RAN model on both classification and sequential tasks, showing its potential for a wide range of applications.

Pretrained language models such as BERT have been successfully applied to a wide range of natural language processing tasks and also achieved impressive performance in document reranking tasks. Recent works indicate that further pretraining the language models on the task-specific datasets before fine-tuning helps improve reranking performance. However, the pre-training tasks like masked language model and next sentence prediction were based on the context of documents instead of encouraging the model to understand the content of queries in document reranking task. In this paper, we propose a new self-supervised joint training framework (SJTF) with a self-supervised method called Masked Query Prediction (MQP) to establish semantic relations between given queries and positive documents. The framework randomly masks a token of query and encodes the masked query paired with positive documents, and uses a linear layer as a decoder to predict the masked token. In addition, the MQP is used to jointly optimize the models with supervised ranking objective during fine-tuning stage without an extra further pre-training stage. Extensive experiments on the MS MARCO passage ranking and TREC Robust datasets show that models trained with our framework obtain significant improvements compared to original models.

Mixed counting models that use the negative binomial distribution as the prior can well model over-dispersed and hierarchically dependent random variables; thus they have attracted much attention in mining dispersed document topics. However, the existing parameter inference method like Monte Carlo sampling is quite time-consuming. In this paper, we propose two efficient neural mixed counting models, i.e., the Negative Binomial-Neural Topic Model (NB-NTM) and the Gamma Negative Binomial-Neural Topic Model (GNB-NTM) for dispersed topic discovery. Neural variational inference algorithms are developed to infer model parameters by using the reparameterization of Gamma distribution and the Gaussian approximation of Poisson distribution. Experiments on real-world datasets indicate that our models outperform state-of-the-art baseline models in terms of perplexity and topic coherence. The results also validate that both NB-NTM and GNB-NTM can produce explainable intermediate variables by generating dispersed proportions of document topics.

Label-specific topics can be widely used for supporting personality psychology, aspect-level sentiment analysis, and cross-domain sentiment classification. To generate label-specific topics, several supervised topic models which adopt likelihood-driven objective functions have been proposed. However, it is hard for them to get a precise estimation on both topic discovery and supervised learning. In this study, we propose a supervised topic model based on the Siamese network, which can trade off label-specific word distributions with document-specific label distributions in a uniform framework. Experiments on real-world datasets validate that our model performs competitive in topic discovery quantitatively and qualitatively. Furthermore, the proposed model can effectively predict categorical or real-valued labels for new documents by generating word embeddings from a label-specific topical space.

This paper focuses on the task of noisy label aggregation in social media, where users with different social or culture backgrounds may annotate invalid or malicious tags for documents. To aggregate noisy labels at a small cost, a network framework is proposed by calculating the matching degree of a document’s topics and the annotators’ meta-data. Unlike using the back-propagation algorithm, a probabilistic inference approach is adopted to estimate network parameters. Finally, a new simulation method is designed for validating the effectiveness of the proposed framework in aggregating noisy labels.

As Chinese is an ideographic character-based language, the words in the texts are not delimited by spaces. Indexing of Chinese documents is impossible without a proper segmentation algorithm. Many Chinese segmentation algorithms have been proposed in the past. Traditional segmentation algorithms cannot operate without a large dictionary or a large corpus of training data. Nowadays, the Web has become the largest corpus that is ideal for Chinese segmentation. Although the search engines do not segment texts into proper words, they maintain huge databases of documents and frequencies of character sequences in the documents. Their databases are important potential resources for segmentation. In this paper, we propose a segmentation algorithm by mining web data with the help from search engines. It is the first unified segmentation algorithm for Chinese language from different geographical areas. Experiments have been conducted on the datasets of a recent Chinese segmentation competition. The results show that our algorithm outperforms the traditional algorithms in terms of precision and recall. Moreover, our algorithm can effectively deal with the problem of segmentation ambiguity, new word (unknown word) detection, and stop words.

In modern information retrieval systems, effective indexing can be achieved by removal of stop words. Till now many stop word lists have been developed for English language. However, no standard stop word list has been constructed for Chinese language yet. With the fast development of information retrieval in Chinese language, exploring the evaluation of Chinese stop word lists becomes critical. In this paper, to save the time and release the burden of manual comparison, we propose a novel stop word list evaluation method with a mutual information-based Chinese segmentation methodology. Experiments have been conducted on training texts taken from a recent international Chinese segmentation competition. Results show that effective stop word lists can improve the accuracy of Chinese segmentation significantly.