Multiple-choice question answering (MCQA) requires the model to select the correct answer from a set of candidate options when given a passage and a question. Previous research has achieved promising results with the assistance of Pre-trained Language Models(PrLMs). However, it has been observed that these models heavily rely on text similarity for inference. In this study, we approach the MCQA task from a causal perspective, treating text similarity as a confounder and establishing a structural causal model(SCM) for the MCQA task. To mitigate the impact of this confounder, we propose a novel Causal Dual Multi-head Co-Attention (CDUMA) model. CDUMA reduces the influence of text similarity between questions and options on model inference and enhances the model’s generalization ability during routine testing. Experimental results on two widely used datasets demonstrate the effectiveness of our approach.

Mitigating linguistic bias and attaining modal equilibrium in Visual Question Answering (VQA) tasks constitute a pivotal concern. Previous work has mainly focused on data augmentation or a uni-modal approach, which is insufficient to fully utilize bi-modal information. In this work, we propose a new causal modal equilibrium framework CausalME, addressing the issue from a causal perspective. CausalME utilizes a question-only branch to capture the linguistic bias of the textual modality and mitigate its causal effect with a newly designed adaptive paradigm. Additionally, CausalME employs counterfactual generation to enhance the causal effect of visual modality. By optimizing the objective function of the entire VQA model, CausalME balances the causal effects of bi-modalities and explicitly guides the model to align text and image information. We conducted extensive experiments and the results show that CausalME brings significant improvements and achieves competitive performance on the bias-sensitive VQA-CP v2 dataset.

Despite the remarkable advancements in Visual Question Answering (VQA), the challenge of mitigating the language bias introduced by textual information remains unresolved. Previous approaches capture language bias from a coarse-grained perspective. However, the finer-grained information within a sentence, such as context and keywords, can result in different biases. Due to the ignorance of fine-grained information, most existing methods fail to sufficiently capture language bias. In this paper, we propose a novel causal intervention training scheme named CIBi to eliminate language bias from a finer-grained perspective. Specifically, we divide the language bias into context bias and keyword bias. We employ causal intervention and contrastive learning to eliminate context bias and improve the multi-modal representation. Additionally, we design a new question-only branch based on counterfactual generation to distill and eliminate keyword bias. Experimental results illustrate that CIBi is applicable to various VQA models, yielding competitive performance.

Numerous studies have been conducted on estimating blood pressure (BP) through photoplethysmogram (PPG) waveform. However, the PPG pulses are generated in the arterial vessel approximately every second, and the morphological features of the continuous pulses are similar. The presence of repeated pulses can lead to over-training in BP prediction. This study proposes a clustering algorithm that to classify PPG pulses into specific clusters using PPG frequency domain features. In this study, the PPG pulses are processed using fast Fourier transform. Subsequently, the PPG frequency, amplitude, phase, real part, and imaginary part of the first to fourth frequency components are extracted. In the proposed K-Means algorithm, a novel distance function is developed to assess the dissimilarity between two pulses. In addition, a cluster center merging framework is proposed to overcome the challenge of selecting the K-value. The experiment utilized in the Medical Information Mart for IntensiveCare II (MIMIC-II) dataset, all the pulses are clustered and merged into 17 clusters. The results demonstrate that the proposed clustering method successfully identifies distinct PPG pulse clusters corresponding to different BP ranges. This finding supports the notion that PPG pulse shape is correlated with BP and enhances the interpretability of BP estimation based on pulse wave analysis.

The spread of rumors along with breaking events seriously hinders the truth in the era of social media. Previous studies reveal that due to the lack of annotated resources, rumors presented in minority languages are hard to be detected. Furthermore, the unforeseen breaking events not involved in yesterday's news exacerbate the scarcity of data resources. In this work, we propose a novel zero-shot framework based on prompt learning to detect rumors falling in different domains or presented in different languages. More specifically, we firstly represent rumor circulated on social media as diverse propagation threads, then design a hierarchical prompt encoding mechanism to learn language-agnostic contextual representations for both prompts and rumor data. To further enhance domain adaptation, we model the domain-invariant structural features from the propagation threads, to incorporate structural position representations of influential community response. In addition, a new virtual response augmentation method is used to improve model training. Extensive experiments conducted on three real-world datasets demonstrate that our proposed model achieves much better performance than state-of-the-art methods and exhibits a superior capacity for detecting rumors at early stages.

Machine Reading Comprehension (MRC) is a challenging task in natural language understanding, especially multi-hop question answering (QA) in long texts. One of the challenges in multi-hop QA requires models to produce interpretable answers based on evidence that is selected from a given long text. Based on the Retriever-Reader architecture, existing work tackles this problem by using different methods to exploit various evidence information. To better use evidence information, we propose a loss function considering answer groups, which improves the reasoning ability of the reader in the Retriever-Reader architecture. Besides, we introduce the relevance constraint factor containing evidence information to improve the reader’s ability of locating key sentences. Evaluated on the HotpotQA dataset, the proposed methods achieve improvement, demonstrating the effectiveness of our methods and the importance of evidence information.

Multiple-choice question answering (MCQA) is a challenging task that requires selecting the correct answer from a set of options based on a given question. There is a trend to use pre-trained encoder-decoder models to solve MCQA. Previous works concentrate on the decoder and adopt the generated text to enhance model performance. However, few studies have optimized the use of encoders for the characteristics of MCQA. In this work, we propose a dynamic exclusion model for MCQA named ExcMC, which mimics human thinking in selection. It dynamically eliminates several incorrect options to optimize the encoder usage. ExcMC outperforms existing comparable works on two widely-used MCQA datasets, demonstrating the effectiveness of our model.

Visual Question Answering (VQA) is a challenging task which requires a fine-grained semantic understanding of visual and textual contents. Existing works focus on better modality representations. However, these methods give little consideration to the long-tailed data distribution in common VQA datasets. The extreme class imbalance causes training bias to behave well in head class, but fail in tail class. Therefore, we propose a unified Adaptive Re-balancing Network (ARN) to take care of classification in both head and tail classes, exhaustively improving performance for VQA. Specifically, two training branches are introduced to per-form their own duty iteratively, which learn the universal representations first and then emphasize the tail data progressively by the re-balancing branch with adaptive learning. Meanwhile, contextual information in the question is vital for guiding accurate visual attention. Thus our network is further equipped with a novel gate mechanism to give higher weight to contextual information. The Experimental results on common benchmarks such as VQA-v2 have demonstrated the superiority of our method compared with state of the art.

Multi-label text classification is an essential task in natural language processing. Existing multi-label classification models generally consider labels as categorical variables and ignore the exploitation of label semantics. In this paper, we view the task as a correlation-guided text representation problem: an attention-based two-step framework is proposed to integrate text information and label semantics by jointly learning words and labels in the same space. In this way, we aim to capture high-order label-label correlations as well as context-label correlations. Specifically, the proposed approach works by learning token-level representations of words and labels globally through a multi-layer Transformer and constructing an attention vector through word-label correlation matrix to generate the text representation. It ensures that relevant words receive higher weights than irrelevant words and thus directly optimizes the classification performance. Extensive experiments over benchmark multi-label datasets clearly validate the effectiveness of the proposed approach, and further analysis demonstrates that it is competitive in both predicting low-frequency labels and convergence speed.

We propose a multi-granularity co-reasoning model for multi-choice reading comprehension task, which aims to select the correct option based on the interaction between passage, question and candidate options. Firstly, we introduce a multiple granularity text matching module to interact passage with question and each option. We take advantage of information extracted from diverse semantic spaces to conduct more extensive matching between text sequences. With this help, we could better match the passage against the question and each option to gather relevant information. Furthermore, we employ a multi-sentence co-reasoning module for sentence inference across multiple sentences. Specifically, we utilize 1D Convolutional Neural Network (1D-CNN) with different kernel sizes and self-attentive Recurrent Neural Network (RNN) to model the relationships of relevant sentences. This module could better synthesize and aggregate sentence-level evidence to make decisions. Experimental results demonstrate that our proposed model achieves state-of-the-art performance for single models on the RACE dataset.

In machine reading comprehension (MRC) tasks, sentence inference is an important but extremely difficult problem. Most of MRC models directly interact articles with questions from the word level, which ignores inter and intra information of sentences and cannot well focus on problems about sentence reasoning and inference, especially when the answer clues are far apart in the article. In this paper, we propose an option gate approach for reading comprehension. We consider applying a sentence-level option gate module to make the model incorporate sentence information. In our approach we (1) extract key sentences in the article to filter out noise unrelated to the question and the options, (2) encode each sentence in articles, questions and options with dot-product self-attention to obtain intra sentence representations, (3) model inter relationships between the article and the question with bilinear attention and (4) apply an option gate with sentence inference information to each option representation with the question-aware article representation. This module can help better reasoning instead of directly word matching or paraphrasing. And this module can easily supply sentence information for most of the existing reading comprehension models. Experimental results on the RACE dataset show that this easy and simple module helps outperform the baseline models by 2.5% at most (single model), and achieve state-of-the-art results on the RACE-H dataset.