Open-domain dialogue generation has gained increasing attention in Natural Language Processing. Its evaluation requires a holistic means. Human ratings are deemed as the gold standard. As human evaluation is inefficient and costly, an automated substitute is highly desirable. In this paper, we propose holistic evaluation metrics that capture different aspects of open-domain dialogues. Our metrics consist of (1) GPT-2 based context coherence between sentences in a dialogue, (2) GPT-2 based fluency in phrasing, (3) n-gram based diversity in responses to augmented queries, and (4) textual-entailment-inference based logical self-consistency. The empirical validity of our metrics is demonstrated by strong correlations with human judgments. We open source the code and relevant materials. 1 * Equal contributions.

Unsupervised dependency parsing aims to learn a dependency grammar from text annotated with only POS tags. Various features and inductive biases are often used to incorporate prior knowledge into learning. One useful type of prior information is that there exist correlations between the parameters of grammar rules involving different POS tags. Previous work employed manually designed features or special prior distributions to encode such information. In this paper, we propose a novel approach to unsupervised dependency parsing that uses a neural model to predict grammar rule probabilities based on distributed representation of POS tags. The distributed representation is automatically learned from data and captures the correlations between POS tags. Our experiments show that our approach outperforms previous approaches utilizing POS correlations and is competitive with recent state-of-the-art approaches on nine different languages.

We introduce a novel approach named unambiguity regularization for unsupervised learning of probabilistic natural language grammars. The approach is based on the observation that natural language is remarkably unambiguous in the sense that only a tiny portion of the large number of possible parses of a natural language sentence are syntactically valid. We incorporate an inductive bias into grammar learning in favor of grammars that lead to unambiguous parses on natural language sentences. The resulting family of algorithms includes the expectation-maximization algorithm (EM) and its variant, Viterbi EM, as well as a so-called softmax-EM algorithm. The softmax-EM algorithm can be implemented with a simple and computationally efficient extension to standard EM. In our experiments of unsupervised dependency grammar learning, we show that unambiguity regularization is beneficial to learning, and in combination with annealing (of the regularization strength) and sparsity priors it leads to improvement over the current state of the art.

Unsupervised dependency parsing, which tries to discover linguistic dependency structures from unannotated data, is a very challenging task. Almost all previous work on this task focuses on learning generative models. In this paper, we develop an unsupervised dependency parsing model based on the CRF autoencoder. The encoder part of our model is discriminative and globally normalized which allows us to use rich features as well as universal linguistic priors. We propose an exact algorithm for parsing as well as a tractable learning algorithm. We evaluated the performance of our model on eight multilingual treebanks and found that our model achieved comparable performance with state-of-the-art approaches.

Visual attention, which assigns weights to image regions according to their relevance to a question, is considered as an indispensable part by most Visual Question Answering models. Although the questions may involve complex relations among multiple regions, few attention models can effectively encode such cross-region relations. In this paper, we demonstrate the importance of encoding such relations by showing the limited effective receptive field of ResNet on two datasets, and propose to model the visual attention as a multivariate distribution over a grid-structured Conditional Random Field on image regions. We demonstrate how to convert the iterative inference algorithms, Mean Field and Loopy Belief Propagation, as recurrent layers of an end-to-end neural network. We empirically evaluated our model on 3 datasets, in which it surpasses the best baseline model of the newly released CLEVR dataset [13] by 9.5%, and the best published model on the VQA dataset [3] by 1.25%. Source code is available at https: //github.com/zhuchen03/vqa-sva.