While there has been substantial progress in text comprehension through simple factoid question answering, more holistic comprehension of a discourse still presents a major challenge (Dunietz et al., 2020).Someone critically reflecting on a text as they read it will pose curiosity-driven, often open-ended questions, which reflect deep understanding of the content and require complex reasoning to answer (Ko et al., 2020;Westera et al., 2020).A key challenge in building and evaluating models for this type of discourse comprehension is the lack of annotated data, especially since collecting answers to such questions requires high cognitive load for annotators.This paper presents a novel paradigm that enables scalable data collection targeting the comprehension of news documents, viewing these questions through the lens of discourse.The resulting corpus, DCQA (Discourse Comprehension by Question Answering), captures both discourse and semantic links between sentences in the form of free-form, open-ended questions.On an evaluation set that we annotated on questions from Ko et al. (2020), we show that DCQA provides valuable supervision for answering openended questions.We additionally design pretraining methods utilizing existing questionanswering resources, and use synthetic data to accommodate unanswerable questions.

Wei-Jen Ko, Ahmed El-Kishky, Adithya Renduchintala, Vishrav Chaudhary, Naman Goyal, Francisco Guzmán, Pascale Fung, Philipp Koehn, Mona Diab. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Inquisitive probing questions come naturally to humans in a variety of settings, but is a challenging task for automatic systems.One natural type of question to ask tries to fill a gap in knowledge during text comprehension, like reading a news article: we might ask about background information, deeper reasons behind things occurring, or more.Despite recent progress with data-driven approaches, generating such questions is beyond the range of models trained on existing datasets.We introduce INQUISITIVE, a dataset of ∼19K questions that are elicited while a person is reading through a document.Compared to existing datasets, INQUISITIVE questions target more towards high-level (semantic and discourse) comprehension of text.We show that readers engage in a series of pragmatic strategies to seek information.Finally, we evaluate question generation models based on GPT-2 (Radford et al., 2019) and show that our model is able to generate reasonable questions although the task is challenging, and highlight the importance of context to generate INQUIS-ITIVE questions.

Existing open-domain dialogue generation models are usually trained to mimic the gold response in the training set using cross-entropy loss on the vocabulary.However, a good response does not need to resemble the gold response, since there are multiple possible responses to a given prompt.In this work, we hypothesize that the current models are unable to integrate information from multiple semantically similar valid responses of a prompt, resulting in the generation of generic and uninformative responses.To address this issue, we propose an alternative to the end-to-end classification on vocabulary.We learn the pair relationship between the prompts and responses as a regression task on a latent space instead.In our novel dialog generation model, the representations of semantically related sentences are close to each other on the latent space.Human evaluation showed that learning the task on a continuous space can generate responses that are both relevant and informative.

Recent advances in NLP have been attributed to the emergence of large-scale pre-trained language models.GPT-2 (Radford et al., 2019), in particular, is suited for generation tasks given its left-to-right language modeling objective, yet the linguistic quality of its generated text has largely remain unexplored.Our work takes a step in understanding GPT-2's outputs in terms of discourse coherence.We perform a comprehensive study on the validity of explicit discourse relations in GPT-2's outputs under both organic generation and fine-tuned scenarios.Results show GPT-2 does not always generate text containing valid discourse relations; nevertheless, its text is more aligned with human expectation in the fine-tuned scenario.We propose a decoupled strategy to mitigate these problems and highlight the importance of explicitly modeling discourse information.

Sentence specificity quantifies the level of detail in a sentence, characterizing the organization of information in discourse. While this information is useful for many downstream applications, specificity prediction systems predict very coarse labels (binary or ternary) and are trained on and tailored toward specific domains (e.g., news). The goal of this work is to generalize specificity prediction to domains where no labeled data is available and output more nuanced realvalued specificity ratings.We present an unsupervised domain adaptation system for sentence specificity prediction, specifically designed to output real-valued estimates from binary training labels. To calibrate the values of these predictions appropriately, we regularize the posterior distribution of the labels towards a reference distribution. We show that our framework generalizes well to three different domains with 50%-68% mean absolute error reduction than the current state-of-the-art system trained for news sentence specificity. We also demonstrate the potential of our work in improving the quality and informativeness of dialogue generation systems.

Multi-label classification is a practical yet challenging task in machine learning related fields, since it requires the prediction of more than one label category for each input instance. We propose a novel deep neural networks (DNN) based model, Canonical Correlated AutoEncoder (C2AE), for solving this task. Aiming at better relating feature and label domain data for improved classification, we uniquely perform joint feature and label embedding by deriving a deep latent space, followed by the introduction of label-correlation sensitive loss function for recovering the predicted label outputs. Our C2AE is achieved by integrating the DNN architectures of canonical correlation analysis and autoencoder, which allows end-to-end learning and prediction with the ability to exploit label dependency. Moreover, our C2AE can be easily extended to address the learning problem with missing labels. Our experiments on multiple datasets with different scales confirm the effectiveness and robustness of our proposed method, which is shown to perform favorably against state-of-the-art methods for multi-label classification.

It is crucial for language models to model long-term dependency in word sequences, which can be achieved to some good extent by recurrent neural network (RNN) based language models with long short-term memory (LSTM) units. To accurately model the sophisticated long-term information in human languages, large memory in language models is necessary. However, the size of RNN-based language models cannot be arbitrarily increased because the computational resources required and the model complexity will also be increase accordingly, due to the limitation of the structure. To overcome this problem, inspired from Neural Turing Machine and Memory Network, we equip RNN-based language models with controllable external memory. With a learnable memory controller, the size of the external memory is independent to the number of model parameters, so the proposed language model can have larger memory without increasing the parameters. In the experiments, the proposed model yielded lower perplexities than RNN-based language models with LSTM units on both English and Chinese corpora.

Real-world visual classification tasks typically need to deal with data observed from different domains. Inspired by canonical correlation analysis (CCA), we propose an enhanced CCA with local density for associating and recognizing cross-domain data. In addition to maximizing the correlation of the projected cross-domain data, our CCA model further exploits the local density information observed from each domain. As a result, our CCA not only exhibits excellent abilities in identifying representative data, noisy data like outliers can be further suppressed during the derivation of our CCA subspace. In our experiments, we successfully apply the proposed methods for solving two cross-domain classification tasks: person re-identification and cross-view action recognition.

Face hallucination technique generates high-resolution face images from low-resolution ones. In this paper, we propose a patch based multitask deep learning method for face hallucination, which is robust to blurring of images. Our method is based on fully connected feedforward neural network, and the weights of the final layers are fine-tuned separately on different clusters of patches. Experimental results show that our system outperforms the prior state-of-the-art methods by a significant margin, while using less testing computation time.

With the goal of increasing the resolution of face images, recent face hallucination methods advance learning techniques which observe training low and high-resolution patches for recovering the output image of interest. Since most existing patch-based face hallucination approaches do not consider the location information of the patches to be hallucinated, the resulting performance might be limited. In this paper, we propose an anchored patch-based hallucination method, which is able to exploit and identify image patches exhibiting structurally and spatially similar information. With these representative anchors observed, improved performance and computation efficiency can be achieved. Experimental results demonstrate that our proposed method achieves satisfactory performance and performs favorably against recent face hallucination approaches.