Answering questions in the context of videos can be helpful in video indexing, video retrieval systems, video summarization, learning management systems and surveillance video analysis. Although there exists a large body of work on visual question answering, work on video question answering (1) is limited to domains like movies, TV shows, gameplay, or human activity, and (2) is mostly based on common sense reasoning. In this paper, we explore a new frontier in video question answering: answering knowledge-based questions in the context of news videos. To this end, we curate a new dataset of 12K news videos spanning across 156 hours with 1M multiple-choice question-answer pairs covering 8263 unique entities. We make the dataset publicly available. Using this dataset, we propose a novel approach, NEWSKVQA (Knowledge-Aware News Video Question Answering) which performs multi-modal inferencing over textual multiple-choice questions, videos, their transcripts and knowledge base, and presents a strong baseline.
Multiple critical scenarios (like Wikipedia text generation given English Infoboxes) need automated generation of descriptive text in low resource (LR) languages from English fact triples. Previous work has focused on English fact-to-text (F2T) generation. To the best of our knowledge, there has been no previous attempt on cross-lingual alignment or generation for LR languages. Building an effective cross-lingual F2T (XF2T) system requires alignment between English structured facts and LR sentences. We propose two unsupervised methods for cross-lingual alignment. We contribute XALIGN, an XF2T dataset with 0.45M pairs across 8 languages, of which 5402 pairs have been manually annotated. We also train strong baseline XF2T generation models on the XAlign dataset.
Although many pretrained models exist for text or images, there have been relatively fewer attempts to train representations specifically for dialog understanding. Prior works usually relied on finetuned representations based on generic text representation models like BERT or GPT-2. But, existing pretraining objectives do not take the structural information of text into consideration. Although generative dialog models can learn structural features too, we argue that the structure-unaware word-by-word generation is not suitable for effective conversation modeling. We empirically demonstrate that such representations do not perform consistently across various dialog understanding tasks. Hence, we propose a structure-aware Mutual Information based loss-function DMI (Discourse Mutual Information) for training dialog-representation models, that additionally captures the inherent uncertainty in response prediction. Extensive evaluation on nine diverse dialog modeling tasks shows that our proposed DMI-based models outperform strong baselines by significant margins, even with small-scale pretraining. Our models show the most promising performance on the dialog evaluation task DailyDialog++, in both random and adversarial negative scenarios.
Coherence is an important aspect of text quality and is crucial for ensuring its readability. It is essential desirable for outputs from text generation systems like summarization, question answering, machine translation, question generation, table-to-text, etc. An automated coherence scoring model is also helpful in essay scoring or providing writing feedback. A large body of previous work has leveraged entity-based methods, syntactic patterns, discourse relations, and more recently traditional deep learning architectures for text coherence assessment. Previous work suffers from drawbacks like the inability to handle long-range dependencies, out-of-vocabulary words, or model sequence information. We hypothesize that coherence assessment is a cognitively complex task that requires deeper models and can benefit from other related tasks. Accordingly, in this paper, we propose four different Transformer-based architectures for the task: vanilla Transformer, hierarchical Transformer, multi-task learning-based model, and a model with fact-based input representation. Our experiments with popular benchmark datasets across multiple domains on four different coherence assessment tasks demonstrate that our models achieve state-of-the-art results outperforming existing models by a good margin.
Deep networks often make confident, yet incorrect, predictions when tested with outlier data that is far removed from their training distributions. Likelihoods computed by deep generative models are a candidate metric for outlier detection with unlabeled data. Yet, previous studies have shown that such likelihoods are unreliable and can be easily biased by simple transformations to input data. Here, we examine outlier detection with variational autoencoders (VAEs), among the simplest class of deep generative models. First, we show that a theoretically-grounded correction readily ameliorates a key bias with VAE likelihood estimates. The bias correction is model-free, sample-specific, and accurately computed with the Bernoulli and continuous Bernoulli visible distributions. Second, we show that a well-known preprocessing technique, contrast normalization, extends the effectiveness of bias correction to natural image datasets. Third, we show that the variance of the likelihoods computed over an ensemble of VAEs also enables robust outlier detection. We perform a comprehensive evaluation of our remedies with nine (grayscale and natural) image datasets, and demonstrate significant advantages, in terms of both speed and accuracy, over four other state-of-the-art methods. Our lightweight remedies are biologically inspired and may serve to achieve efficient outlier detection with many types of deep generative models.
The exponential rise of online social media has enabled the creation, distribution, and consumption of information at an unprecedented rate. However, it has also led to the burgeoning of various forms of online abuse. Increasing cases of online antisemitism have become one of the major concerns because of its socio-political consequences. Unlike other major forms of online abuse like racism, sexism, etc., online antisemitism has not been studied much from a machine learning perspective. To the best of our knowledge, we present the first work in the direction of automated multimodal detection of online antisemitism. The task poses multiple challenges that include extracting signals across multiple modalities, contextual references, and handling multiple aspects of antisemitism. Unfortunately, there does not exist any publicly available benchmark corpus for this critical task. Hence, we collect and label two datasets with 3,102 and 3,509 social media posts from Twitter and Gab respectively. Further, we present a multimodal deep learning system that detects the presence of antisemitic content and its specific antisemitism category using text and images from posts. We perform an extensive set of experiments on the two datasets to evaluate the efficacy of the proposed system. Finally, we also present a qualitative analysis of our study.
Automatic text summarization has been widely studied as an important task in natural language processing. Traditionally, various feature engineering and machine learning based systems have been proposed for extractive as well as abstractive text summarization. Recently, deep learning based, specifically Transformer-based systems have been immensely popular. Summarization is a cognitively challenging task - extracting summary worthy sentences is laborious, and expressing semantics in brief when doing abstractive summarization is complicated. In this paper, we specifically look at the problem of summarizing scientific research papers from multiple domains. We differentiate between two types of summaries, namely, (a) LaySumm: A very short summary that captures the essence of the research paper in layman terms restricting overtly specific technical jargon and (b) LongSumm: A much longer detailed summary aimed at providing specific insights into various ideas touched upon in the paper. While leveraging latest Transformer-based models, our systems are simple, intuitive and based on how specific paper sections contribute to human summaries of the two types described above. Evaluations against gold standard summaries using ROUGE metrics prove the effectiveness of our approach. On blind test corpora, our system ranks first and third for the LongSumm and LaySumm tasks respectively.
Although several automatic itinerary generation services have made travel planning easy, often times travellers find themselves in unique situations where they cannot make the best out of their trip. Visitors differ in terms of many factors such as suffering from a disability, being of a particular dietary preference, travelling with a toddler, etc. While most tourist spots are universal, others may not be inclusive for all. In this paper, we focus on the problem of mining inclusion and exclusion phrases associated with 11 such factors, from reviews related to a tourist spot. While existing work on tourism data mining mainly focuses on structured extraction of trip related information, personalized sentiment analysis, and automatic itinerary generation, to the best of our knowledge this is the first work on inclusion/exclusion phrase mining from tourism reviews. Using a dataset of 2000 reviews related to 1000 tourist spots, our broad level classifier provides a binary overlap F1 of $\sim$80 and $\sim$82 to classify a phrase as inclusion or exclusion respectively. Further, our inclusion/exclusion classifier provides an F1 of $\sim$98 and $\sim$97 for 11-class inclusion and exclusion classification respectively. We believe that our work can significantly improve the quality of an automatic itinerary generation service.
While extensive popularity of online social media platforms has made information dissemination faster, it has also resulted in widespread online abuse of different types like hate speech, offensive language, sexist and racist opinions, etc. Detection and curtailment of such abusive content is critical for avoiding its psychological impact on victim communities, and thereby preventing hate crimes. Previous works have focused on classifying user posts into various forms of abusive behavior. But there has hardly been any focus on estimating the severity of abuse and the target. In this paper, we present a first of the kind dataset with 7601 posts from Gab which looks at online abuse from the perspective of presence of abuse, severity and target of abusive behavior. We also propose a system to address these tasks, obtaining an accuracy of ~80% for abuse presence, ~82% for abuse target prediction, and ~65% for abuse severity prediction.