Unlike extractive summarization, abstractive summarization has to fuse different parts of the source text, which inclines to create fake facts. Our preliminary study reveals nearly 30% of the outputs from a state-of-the-art neural summarization system suffer from this problem. While previous abstractive summarization approaches usually focus on the improvement of informativeness, we argue that faithfulness is also a vital prerequisite for a practical abstractive summarization system. To avoid generating fake facts in a summary, we leverage open information extraction and dependency parse technologies to extract actual fact descriptions from the source text. The dual-attention sequence-to-sequence framework is then proposed to force the generation conditioned on both the source text and the extracted fact descriptions. Experiments on the Gigaword benchmark dataset demonstrate that our model can greatly reduce fake summaries by 80%. Notably, the fact descriptions also bring significant improvement on informativeness since they often condense the meaning of the source text.

Current semantic parsers either compute shallow representations over a wide range of input, or deeper representations in very limited domains. We describe a system that provides broad-coverage, deep semantic parsing designed to work in any domain using a core domain-general lexicon, ontology and grammar. This paper discusses how this core system can be customized for a particularly challenging domain, namely reading research papers in biology. We evaluate these customizations with some ablation experiments

Creating systems that can learn to answer natural language questions has been a longstanding challenge for artificial intelligence. Most prior approaches focused on producing a specialized language system for a particular domain and dataset, and they required training on a large corpus manually annotated with logical forms. This paper introduces an analogy-based approach that instead adapts an existing general purpose semantic parser to answer questions in a novel domain by jointly learning disambiguation heuristics and query construction templates from purely textual question-answer pairs. Our technique uses possible semantic interpretations of the natural language questions and answers to constrain a query-generation procedure, producing cases during training that are subsequently reused via analogical retrieval and composed to answer test questions. Bootstrapping an existing semantic parser in this way significantly reduces the number of training examples needed to accurately answer questions. We demonstrate the efficacy of our technique using the Geoquery corpus, on which it approaches state of the art performance using 10-fold cross validation, shows little decrease in performance with 2-folds, and achieves above 50% accuracy with as few as 10 examples.

Many vision and language tasks require commonsense reasoning beyond data-driven image and natural language processing. Here we adopt Visual Question Answering (VQA) as an example task, where a system is expected to answer a question in natural language about an image. Current state-of-the-art systems attempted to solve the task using deep neural architectures and achieved promising performance. However, the resulting systems are generally opaque and they struggle in understanding questions for which extra knowledge is required. In this paper, we present an explicit reasoning layer on top of a set of penultimate neural network based systems. The reasoning layer enables reasoning and answering questions where additional knowledge is required, and at the same time provides an interpretable interface to the end users. Specifically, the reasoning layer adopts a Probabilistic Soft Logic (PSL) based engine to reason over a basket of inputs: visual relations, the semantic parse of the question, and background ontological knowledge from word2vec and ConceptNet. Experimental analysis of the answers and the key evidential predicates generated on the VQA dataset validate our approach.

We describe MultiSynth, a framework for synthesizing domain-specific programs from a multimodal dataset of examples. Given a domain-specific language (DSL), a dataset is multimodal if there is no single program in the DSL that generalizes over all the examples. Further, even if the examples in the dataset were generalized in terms of a set of programs, the domains of these programs may not be disjoint, thereby leading to ambiguity in synthesis. MultiSynth is a framework that incorporates concepts of synthesizing programs with minimum generality, while addressing the need of accurate prediction. We show how these can be achieved through (i) transformation driven partitioning of the dataset, (ii) least general generalization, for a generalized specification of the input and the output, and (iii) learning to rank, for estimating feature weights in order to map an input to the most appropriate mode in case of ambiguity. We show the effectiveness of our framework in two domains: in the first case, we extend an existing approach for synthesizing programs for XML tree transformations to ambiguous multimodal datasets. In the second case, MultiSynth is used to preorder words for machine translation, by learning permutations of productions in the parse trees of the source side sentences. Our evaluations reflect the effectiveness of our approach.

In this paper, we focus on learning structure-aware document representations from data without recourse to a discourse parser or additional annotations. Drawing inspiration from recent efforts to empower neural networks with a structural bias (Cheng et al., 2016; Kim et al., 2017), we propose a model that can encode a document while automatically inducing rich structural dependencies. Specifically, we embed a differentiable non-projective parsing algorithm into a neural model and use attention mechanisms to incorporate the structural biases. Experimental evaluations across different tasks and datasets show that the proposed model achieves state-of-the-art results on document modeling tasks while inducing intermediate structures which are both interpretable and meaningful.

We train multi-task autoencoders on linguistic tasks and analyze the learned hidden sentence representations. The representations change significantly when translation and part-of-speech decoders are added. The more decoders a model employs, the better it clusters sentences according to their syntactic similarity, as the representation space becomes less entangled. We explore the structure of the representation space by interpolating between sentences, which yields interesting pseudo-English sentences, many of which have recognizable syntactic structure. Lastly, we point out an interesting property of our models: The difference-vector between two sentences can be added to change a third sentence with similar features in a meaningful way.

Staying on top of recent work is an important part of being a good researcher, but this can be quite difficult. Thousands of new papers are published every year at the main ML and NLP conferences, not to mention all the specialised workshops and everything that shows up on ArXiv. Going through all of them, even just to find the papers that you want to read in more depth, can be very time-consuming. In this post, I have summarised 50 papers. After going through a paper, if I had the chance, I would write down a few notes and summarise the work in a couple of sentences. These are not meant as reviews – I'm not commenting on whether I think the paper is good or not. But I do try to present the crux of the paper as bluntly as possible, without unnecessary sales tactics. Hopefully this can give you the general idea of 50 papers, in roughly 20 minutes of reading time. The papers are not selected or ordered based on any criteria. It is not a list of the best papers I have read, more like a random sample.

Modeling various aspects of language--syntax, semantics, pragmatics, and discourse, among others--by the use of constrained formal-computational systems, just adequate for such modeling, has proved to be an effective research strategy, leading to deep understanding of these aspects, with implications for both machine processing and human processing. This approach enables one to distinguish between the universal and stipulative constraints.

I review current statistical work on syntactic parsing and then consider part-of-speech tagging, which was the first syntactic problem to successfully be attacked by statistical techniques and also serves as a good warm-up for the main topic--statistical parsing. Here, I consider both the simplified case in which the input string is viewed as a string of parts of speech and the more interesting case in which the parser is guided by statistical information about the particular words in the sentence. Finally, I anticipate future research directions. In this example, I adopt the standard abbreviations: s for sentence, np for noun phrase, vp for verb phrase, and det for determiner. It is generally accepted that finding the sort of structure shown in figure 1 is useful in determining the meaning of a sentence.