PyTorchKR Papers

최신 논문

총 4,696편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1702.05821

Post-edit Analysis of Collective Biography Generation

Text generation is increasingly common but often requires manual post-editing where high precision is critical to end users. However, manual editing is expensive so we want to ensure this effort is focused on high-value tasks. And we want to maintain stylistic consistency, a particular challenge in crowd settings. We present a case study, analysing human post-editing in the context of a template-based biography generation system. An edit flow visualisation combined with manual characterisation of edits helps identify and prioritise work for improving end-to-end efficiency and accuracy.

arXiv:1608.04077

Generative Knowledge Transfer for Neural Language Models

In this paper, we propose a generative knowledge transfer technique that trains an RNN based language model (student network) using text and output probabilities generated from a previously trained RNN (teacher network). The text generation can be conducted by either the teacher or the student network. We can also improve the performance by taking the ensemble of soft labels obtained from multiple teacher networks. This method can be used for privacy conscious language model adaptation because no user data is directly used for training. Especially, when the soft labels of multiple devices are aggregated via a trusted third party, we can expect very strong privacy protection.

arXiv:1607.07086공식 코드 · Papers with Code

An Actor-Critic Algorithm for Sequence Prediction

We present an approach to training neural networks to generate sequences using actor-critic methods from reinforcement learning (RL). Current log-likelihood training methods are limited by the discrepancy between their training and testing modes, as models must generate tokens conditioned on their previous guesses rather than the ground-truth tokens. We address this problem by introducing a \textit{critic} network that is trained to predict the value of an output token, given the policy of an \textit{actor} network. This results in a training procedure that is much closer to the test phase, and allows us to directly optimize for a task-specific score such as BLEU. Crucially, since we leverage these techniques in the supervised learning setting rather than the traditional RL setting, we condition the critic network on the ground-truth output. We show that our method leads to improved performance on both a synthetic task, and for German-English machine translation. Our analysis paves the way for such methods to be applied in natural language generation tasks, such as machine translation, caption generation, and dialogue modelling.

arXiv:1703.02031

Random vector generation of a semantic space

We show how random vectors and random projection can be implemented in the usual vector space model to construct a Euclidean semantic space from a French synonym dictionary. We evaluate theoretically the resulting noise and show the experimental distribution of the similarities of terms in a neighborhood according to the choice of parameters. We also show that the Schmidt orthogonalization process is applicable and can be used to separate homonyms with distinct semantic meanings. Neighboring terms are easily arranged into semantically significant clusters which are well suited to the generation of realistic lists of synonyms and to such applications as word selection for automatic text generation. This process, applicable to any language, can easily be extended to collocations, is extremely fast and can be updated in real time, whenever new synonyms are proposed.

arXiv:1703.10429

An Empirical Approach for Modeling Fuzzy Geographical Descriptors

We present a novel heuristic approach that defines fuzzy geographical descriptors using data gathered from a survey with human subjects. The participants were asked to provide graphical interpretations of the descriptors `north' and `south' for the Galician region (Spain). Based on these interpretations, our approach builds fuzzy descriptors that are able to compute membership degrees for geographical locations. We evaluated our approach in terms of efficiency and precision. The fuzzy descriptors are meant to be used as the cornerstones of a geographical referring expression generation algorithm that is able to linguistically characterize geographical locations and regions. This work is also part of a general research effort that intends to establish a methodology which reunites the empirical studies traditionally practiced in data-to-text and the use of fuzzy sets to model imprecision and vagueness in words and expressions for text generation purposes.

arXiv:1704.01696

A Syntactic Neural Model for General-Purpose Code Generation

We consider the problem of parsing natural language descriptions into source code written in a general-purpose programming language like Python. Existing data-driven methods treat this problem as a language generation task without considering the underlying syntax of the target programming language. Informed by previous work in semantic parsing, in this paper we propose a novel neural architecture powered by a grammar model to explicitly capture the target syntax as prior knowledge. Experiments find this an effective way to scale up to generation of complex programs from natural language descriptions, achieving state-of-the-art results that well outperform previous code generation and semantic parsing approaches.

arXiv:1704.00057

Frames: A Corpus for Adding Memory to Goal-Oriented Dialogue Systems

This paper presents the Frames dataset (Frames is available at http://datasets.maluuba.com/Frames), a corpus of 1369 human-human dialogues with an average of 15 turns per dialogue. We developed this dataset to study the role of memory in goal-oriented dialogue systems. Based on Frames, we introduce a task called frame tracking, which extends state tracking to a setting where several states are tracked simultaneously. We propose a baseline model for this task. We show that Frames can also be used to study memory in dialogue management and information presentation through natural language generation.

arXiv:1704.06851

Affect-LM: A Neural Language Model for Customizable Affective Text Generation

Human verbal communication includes affective messages which are conveyed through use of emotionally colored words. There has been a lot of research in this direction but the problem of integrating state-of-the-art neural language models with affective information remains an area ripe for exploration. In this paper, we propose an extension to an LSTM (Long Short-Term Memory) language model for generating conversational text, conditioned on affect categories. Our proposed model, Affect-LM enables us to customize the degree of emotional content in generated sentences through an additional design parameter. Perception studies conducted using Amazon Mechanical Turk show that Affect-LM generates naturally looking emotional sentences without sacrificing grammatical correctness. Affect-LM also learns affect-discriminative word representations, and perplexity experiments show that additional affective information in conversational text can improve language model prediction.

arXiv:1606.03632

Natural Language Generation in Dialogue using Lexicalized and Delexicalized Data

Natural language generation plays a critical role in spoken dialogue systems. We present a new approach to natural language generation for task-oriented dialogue using recurrent neural networks in an encoder-decoder framework. In contrast to previous work, our model uses both lexicalized and delexicalized components i.e. slot-value pairs for dialogue acts, with slots and corresponding values aligned together. This allows our model to learn from all available data including the slot-value pairing, rather than being restricted to delexicalized slots. We show that this helps our model generate more natural sentences with better grammar. We further improve our model's performance by transferring weights learnt from a pretrained sentence auto-encoder. Human evaluation of our best-performing model indicates that it generates sentences which users find more appealing.

arXiv:1704.08531

A Survey of Neural Network Techniques for Feature Extraction from Text

This paper aims to catalyze the discussions about text feature extraction techniques using neural network architectures. The research questions discussed in the paper focus on the state-of-the-art neural network techniques that have proven to be useful tools for language processing, language generation, text classification and other computational linguistics tasks.

arXiv:1704.08798

Word Affect Intensities

Words often convey affect -- emotions, feelings, and attitudes. Further, different words can convey affect to various degrees (intensities). However, existing manually created lexicons for basic emotions (such as anger and fear) indicate only coarse categories of affect association (for example, associated with anger or not associated with anger). Automatic lexicons of affect provide fine degrees of association, but they tend not to be accurate as human-created lexicons. Here, for the first time, we present a manually created affect intensity lexicon with real-valued scores of intensity for four basic emotions: anger, fear, joy, and sadness. (We will subsequently add entries for more emotions such as disgust, anticipation, trust, and surprise.) We refer to this dataset as the NRC Affect Intensity Lexicon, or AIL for short. AIL has entries for close to 6,000 English words. We used a technique called best-worst scaling (BWS) to create the lexicon. BWS improves annotation consistency and obtains reliable fine-grained scores (split-half reliability > 0.91). We also compare the entries in AIL with the entries in the NRC VAD Lexicon, which has valence, arousal, and dominance (VAD) scores for 20K English words. We find that anger, fear, and sadness words, on average, have very similar VAD scores. However, sadness words tend to have slightly lower dominance scores than fear and anger words. The Affect Intensity Lexicon has applications in automatic emotion analysis in a number of domains such as commerce, education, intelligence, and public health. AIL is also useful in the building of natural language generation systems.

arXiv:1702.00500

AMR-to-text Generation with Synchronous Node Replacement Grammar

This paper addresses the task of AMR-to-text generation by leveraging synchronous node replacement grammar. During training, graph-to-string rules are learned using a heuristic extraction algorithm. At test time, a graph transducer is applied to collapse input AMRs and generate output sentences. Evaluated on SemEval-2016 Task 8, our method gives a BLEU score of 25.62, which is the best reported so far.

arXiv:1705.02894

Geometric GAN

Generative Adversarial Nets (GANs) represent an important milestone for effective generative models, which has inspired numerous variants seemingly different from each other. One of the main contributions of this paper is to reveal a unified geometric structure in GAN and its variants. Specifically, we show that the adversarial generative model training can be decomposed into three geometric steps: separating hyperplane search, discriminator parameter update away from the separating hyperplane, and the generator update along the normal vector direction of the separating hyperplane. This geometric intuition reveals the limitations of the existing approaches and leads us to propose a new formulation called geometric GAN using SVM separating hyperplane that maximizes the margin. Our theoretical analysis shows that the geometric GAN converges to a Nash equilibrium between the discriminator and generator. In addition, extensive numerical results show that the superior performance of geometric GAN.

arXiv:1705.03802

Analysing Data-To-Text Generation Benchmarks

Recently, several data-sets associating data to text have been created to train data-to-text surface realisers. It is unclear however to what extent the surface realisation task exercised by these data-sets is linguistically challenging. Do these data-sets provide enough variety to encourage the development of generic, high-quality data-to-text surface realisers ? In this paper, we argue that these data-sets have important drawbacks. We back up our claim using statistics, metrics and manual evaluation. We conclude by eliciting a set of criteria for the creation of a data-to-text benchmark which could help better support the development, evaluation and comparison of linguistically sophisticated data-to-text surface realisers.

arXiv:1706.04033

On Natural Language Generation of Formal Argumentation

In this paper we provide a first analysis of the research questions that arise when dealing with the problem of communicating pieces of formal argumentation through natural language interfaces. It is a generally held opinion that formal models of argumentation naturally capture human argument, and some preliminary studies have focused on justifying this view. Unfortunately, the results are not only inconclusive, but seem to suggest that explaining formal argumentation to humans is a rather articulated task. Graphical models for expressing argumentation-based reasoning are appealing, but often humans require significant training to use these tools effectively. We claim that natural language interfaces to formal argumentation systems offer a real alternative, and may be the way forward for systems that capture human argument.

arXiv:1702.08139

Improved Variational Autoencoders for Text Modeling using Dilated Convolutions

Recent work on generative modeling of text has found that variational auto-encoders (VAE) incorporating LSTM decoders perform worse than simpler LSTM language models (Bowman et al., 2015). This negative result is so far poorly understood, but has been attributed to the propensity of LSTM decoders to ignore conditioning information from the encoder. In this paper, we experiment with a new type of decoder for VAE: a dilated CNN. By changing the decoder's dilation architecture, we control the effective context from previously generated words. In experiments, we find that there is a trade off between the contextual capacity of the decoder and the amount of encoding information used. We show that with the right decoder, VAE can outperform LSTM language models. We demonstrate perplexity gains on two datasets, representing the first positive experimental result on the use VAE for generative modeling of text. Further, we conduct an in-depth investigation of the use of VAE (with our new decoding architecture) for semi-supervised and unsupervised labeling tasks, demonstrating gains over several strong baselines.

arXiv:1706.09433

Data-driven Natural Language Generation: Paving the Road to Success

We argue that there are currently two major bottlenecks to the commercial use of statistical machine learning approaches for natural language generation (NLG): (a) The lack of reliable automatic evaluation metrics for NLG, and (b) The scarcity of high quality in-domain corpora. We address the first problem by thoroughly analysing current evaluation metrics and motivating the need for a new, more reliable metric. The second problem is addressed by presenting a novel framework for developing and evaluating a high quality corpus for NLG training.

arXiv:1707.00251

Where to Play: Retrieval of Video Segments using Natural-Language Queries

In this paper, we propose a new approach for retrieval of video segments using natural language queries. Unlike most previous approaches such as concept-based methods or rule-based structured models, the proposed method uses image captioning model to construct sentential queries for visual information. In detail, our approach exploits multiple captions generated by visual features in each image with `Densecap'. Then, the similarities between captions of adjacent images are calculated, which is used to track semantically similar captions over multiple frames. Besides introducing this novel idea of 'tracking by captioning', the proposed method is one of the first approaches that uses a language generation model learned by neural networks to construct semantic query describing the relations and properties of visual information. To evaluate the effectiveness of our approach, we have created a new evaluation dataset, which contains about 348 segments of scenes in 20 movie-trailers. Through quantitative and qualitative evaluation, we show that our method is effective for retrieval of video segments using natural language queries.

arXiv:1706.09799공식 코드 · Papers with Code

Relevance of Unsupervised Metrics in Task-Oriented Dialogue for Evaluating Natural Language Generation

Automated metrics such as BLEU are widely used in the machine translation literature. They have also been used recently in the dialogue community for evaluating dialogue response generation. However, previous work in dialogue response generation has shown that these metrics do not correlate strongly with human judgment in the non task-oriented dialogue setting. Task-oriented dialogue responses are expressed on narrower domains and exhibit lower diversity. It is thus reasonable to think that these automated metrics would correlate well with human judgment in the task-oriented setting where the generation task consists of translating dialogue acts into a sentence. We conduct an empirical study to confirm whether this is the case. Our findings indicate that these automated metrics have stronger correlation with human judgments in the task-oriented setting compared to what has been observed in the non task-oriented setting. We also observe that these metrics correlate even better for datasets which provide multiple ground truth reference sentences. In addition, we show that some of the currently available corpora for task-oriented language generation can be solved with simple models and advocate for more challenging datasets.

arXiv:1706.09254

The E2E Dataset: New Challenges For End-to-End Generation

This paper describes the E2E data, a new dataset for training end-to-end, data-driven natural language generation systems in the restaurant domain, which is ten times bigger than existing, frequently used datasets in this area. The E2E dataset poses new challenges: (1) its human reference texts show more lexical richness and syntactic variation, including discourse phenomena; (2) generating from this set requires content selection. As such, learning from this dataset promises more natural, varied and less template-like system utterances. We also establish a baseline on this dataset, which illustrates some of the difficulties associated with this data.