The natural language generation (NLG) component of a spoken dialogue system (SDS) usually needs a substantial amount of handcrafting or a well-labeled dataset to be trained on. These limitations add significantly to development costs and make cross-domain, multi-lingual dialogue systems intractable. Moreover, human languages are context-aware. The most natural response should be directly learned from data rather than depending on predefined syntaxes or rules. This paper presents a statistical language generator based on a joint recurrent and convolutional neural network structure which can be trained on dialogue act-utterance pairs without any semantic alignments or predefined grammar trees. Objective metrics suggest that this new model outperforms previous methods under the same experimental conditions. Results of an evaluation by human judges indicate that it produces not only high quality but linguistically varied utterances which are preferred compared to n-gram and rule-based systems.
Natural language generation (NLG) is a critical component of spoken dialogue and it has a significant impact both on usability and perceived quality. Most NLG systems in common use employ rules and heuristics and tend to generate rigid and stylised responses without the natural variation of human language. They are also not easily scaled to systems covering multiple domains and languages. This paper presents a statistical language generator based on a semantically controlled Long Short-term Memory (LSTM) structure. The LSTM generator can learn from unaligned data by jointly optimising sentence planning and surface realisation using a simple cross entropy training criterion, and language variation can be easily achieved by sampling from output candidates. With fewer heuristics, an objective evaluation in two differing test domains showed the proposed method improved performance compared to previous methods. Human judges scored the LSTM system higher on informativeness and naturalness and overall preferred it to the other systems.
Several techniques have been used to generate weather forecast texts. In this paper, case based reasoning (CBR) is proposed for weather forecast text generation because similar weather conditions occur over time and should have similar forecast texts. CBR-METEO, a system for generating weather forecast texts was developed using a generic framework (jCOLIBRI) which provides modules for the standard components of the CBR architecture. The advantage in a CBR approach is that systems can be built in minimal time with far less human effort after initial consultation with experts. The approach depends heavily on the goodness of the retrieval and revision components of the CBR process. We evaluated CBRMETEO with NIST, an automated metric which has been shown to correlate well with human judgements for this domain. The system shows comparable performance with other NLG systems that perform the same task.
The Rational Speech Acts (RSA) model treats language use as a recursive process in which probabilistic speaker and listener agents reason about each other's intentions to enrich the literal semantics of their language along broadly Gricean lines. RSA has been shown to capture many kinds of conversational implicature, but it has been criticized as an unrealistic model of speakers, and it has so far required the manual specification of a semantic lexicon, preventing its use in natural language processing applications that learn lexical knowledge from data. We address these concerns by showing how to define and optimize a trained statistical classifier that uses the intermediate agents of RSA as hidden layers of representation forming a non-linear activation function. This treatment opens up new application domains and new possibilities for learning effectively from data. We validate the model on a referential expression generation task, showing that the best performance is achieved by incorporating features approximating well-established insights about natural language generation into RSA.
The Voynich manuscript is a medieval book written in an unknown script. This paper studies the relation between similarly spelled words in the Voynich manuscript. By means of a detailed analysis of similar spelled words it was possible to reveal the text generation method used for the Voynich manuscript.
Recent language models, especially those based on recurrent neural networks (RNNs), make it possible to generate natural language from a learned probability. Language generation has wide applications including machine translation, summarization, question answering, conversation systems, etc. Existing methods typically learn a joint probability of words conditioned on additional information, which is (either statically or dynamically) fed to RNN's hidden layer. In many applications, we are likely to impose hard constraints on the generated texts, i.e., a particular word must appear in the sentence. Unfortunately, existing approaches could not solve this problem. In this paper, we propose a novel backward and forward language model. Provided a specific word, we use RNNs to generate previous words and future words, either simultaneously or asynchronously, resulting in two model variants. In this way, the given word could appear at any position in the sentence. Experimental results show that the generated texts are comparable to sequential LMs in quality.
Moving from limited-domain natural language generation (NLG) to open domain is difficult because the number of semantic input combinations grows exponentially with the number of domains. Therefore, it is important to leverage existing resources and exploit similarities between domains to facilitate domain adaptation. In this paper, we propose a procedure to train multi-domain, Recurrent Neural Network-based (RNN) language generators via multiple adaptation steps. In this procedure, a model is first trained on counterfeited data synthesised from an out-of-domain dataset, and then fine tuned on a small set of in-domain utterances with a discriminative objective function. Corpus-based evaluation results show that the proposed procedure can achieve competitive performance in terms of BLEU score and slot error rate while significantly reducing the data needed to train generators in new, unseen domains. In subjective testing, human judges confirm that the procedure greatly improves generator performance when only a small amount of data is available in the domain.
In this paper, we present a conversational model that incorporates both context and participant role for two-party conversations. Different architectures are explored for integrating participant role and context information into a Long Short-term Memory (LSTM) language model. The conversational model can function as a language model or a language generation model. Experiments on the Ubuntu Dialog Corpus show that our model can capture multiple turn interaction between participants. The proposed method outperforms a traditional LSTM model as measured by language model perplexity and response ranking. Generated responses show characteristic differences between the two participant roles.
With the recent popularity of animated GIFs on social media, there is need for ways to index them with rich metadata. To advance research on animated GIF understanding, we collected a new dataset, Tumblr GIF (TGIF), with 100K animated GIFs from Tumblr and 120K natural language descriptions obtained via crowdsourcing. The motivation for this work is to develop a testbed for image sequence description systems, where the task is to generate natural language descriptions for animated GIFs or video clips. To ensure a high quality dataset, we developed a series of novel quality controls to validate free-form text input from crowdworkers. We show that there is unambiguous association between visual content and natural language descriptions in our dataset, making it an ideal benchmark for the visual content captioning task. We perform extensive statistical analyses to compare our dataset to existing image and video description datasets. Next, we provide baseline results on the animated GIF description task, using three representative techniques: nearest neighbor, statistical machine translation, and recurrent neural networks. Finally, we show that models fine-tuned from our animated GIF description dataset can be helpful for automatic movie description.
This paper presents an end-to-end neural network model, named Neural Generative Question Answering (GENQA), that can generate answers to simple factoid questions, based on the facts in a knowledge-base. More specifically, the model is built on the encoder-decoder framework for sequence-to-sequence learning, while equipped with the ability to enquire the knowledge-base, and is trained on a corpus of question-answer pairs, with their associated triples in the knowledge-base. Empirical study shows the proposed model can effectively deal with the variations of questions and answers, and generate right and natural answers by referring to the facts in the knowledge-base. The experiment on question answering demonstrates that the proposed model can outperform an embedding-based QA model as well as a neural dialogue model trained on the same data.
We explore the implications of using fuzzy techniques (mainly those commonly used in the linguistic description/summarization of data discipline) from a natural language generation perspective. For this, we provide an extensive discussion of some general convergence points and an exploration of the relationship between the different tasks involved in the standard NLG system pipeline architecture and the most common fuzzy approaches used in linguistic summarization/description of data, such as fuzzy quantified statements, evaluation criteria or aggregation operators. Each individual discussion is illustrated with a related use case. Recent work made in the context of cross-fertilization of both research fields is also referenced. This paper encompasses general ideas that emerged as part of the PhD thesis "Application of fuzzy sets in data-to-text systems". It does not present a specific application or a formal approach, but rather discusses current high-level issues and potential usages of fuzzy sets (focused on linguistic summarization of data) in natural language generation.
Natural Language Generation systems typically have two parts - strategic ('what to say') and tactical ('how to say'). We present our experiments in building an unsupervised corpus-driven template based tactical NLG system. We consider templates as a sequence of words containing gaps. Our idea is based on the observation that templates are grammatical locally (within their textual span). We posit the construction of a sentence as a highly restricted sequence of such templates. This work is an attempt to explore the resulting search space using Genetic Algorithms to arrive at acceptable solutions. We present a baseline implementation of this approach which outputs gapped text.
Many language generation tasks require the production of text conditioned on both structured and unstructured inputs. We present a novel neural network architecture which generates an output sequence conditioned on an arbitrary number of input functions. Crucially, our approach allows both the choice of conditioning context and the granularity of generation, for example characters or tokens, to be marginalised, thus permitting scalable and effective training. Using this framework, we address the problem of generating programming code from a mixed natural language and structured specification. We create two new data sets for this paradigm derived from the collectible trading card games Magic the Gathering and Hearthstone. On these, and a third preexisting corpus, we demonstrate that marginalising multiple predictors allows our model to outperform strong benchmarks.
Recently a variety of LSTM-based conditional language models (LM) have been applied across a range of language generation tasks. In this work we study various model architectures and different ways to represent and aggregate the source information in an end-to-end neural dialogue system framework. A method called snapshot learning is also proposed to facilitate learning from supervised sequential signals by applying a companion cross-entropy objective function to the conditioning vector. The experimental and analytical results demonstrate firstly that competition occurs between the conditioning vector and the LM, and the differing architectures provide different trade-offs between the two. Secondly, the discriminative power and transparency of the conditioning vector is key to providing both model interpretability and better performance. Thirdly, snapshot learning leads to consistent performance improvements independent of which architecture is used.
We introduce the multiresolution recurrent neural network, which extends the sequence-to-sequence framework to model natural language generation as two parallel discrete stochastic processes: a sequence of high-level coarse tokens, and a sequence of natural language tokens. There are many ways to estimate or learn the high-level coarse tokens, but we argue that a simple extraction procedure is sufficient to capture a wealth of high-level discourse semantics. Such procedure allows training the multiresolution recurrent neural network by maximizing the exact joint log-likelihood over both sequences. In contrast to the standard log- likelihood objective w.r.t. natural language tokens (word perplexity), optimizing the joint log-likelihood biases the model towards modeling high-level abstractions. We apply the proposed model to the task of dialogue response generation in two challenging domains: the Ubuntu technical support domain, and Twitter conversations. On Ubuntu, the model outperforms competing approaches by a substantial margin, achieving state-of-the-art results according to both automatic evaluation metrics and a human evaluation study. On Twitter, the model appears to generate more relevant and on-topic responses according to automatic evaluation metrics. Finally, our experiments demonstrate that the proposed model is more adept at overcoming the sparsity of natural language and is better able to capture long-term structure.
We present and evaluate a new model for Natural Language Generation (NLG) in Spoken Dialogue Systems, based on statistical planning, given noisy feedback from the current generation context (e.g. a user and a surface realiser). We study its use in a standard NLG problem: how to present information (in this case a set of search results) to users, given the complex trade- offs between utterance length, amount of information conveyed, and cognitive load. We set these trade-offs by analysing existing MATCH data. We then train a NLG pol- icy using Reinforcement Learning (RL), which adapts its behaviour to noisy feed- back from the current generation context. This policy is compared to several base- lines derived from previous work in this area. The learned policy significantly out- performs all the prior approaches.
Sentence ordering is a general and critical task for natural language generation applications. Previous works have focused on improving its performance in an external, downstream task, such as multi-document summarization. Given its importance, we propose to study it as an isolated task. We collect a large corpus of academic texts, and derive a data driven approach to learn pairwise ordering of sentences, and validate the efficacy with extensive experiments. Source codes and dataset of this paper will be made publicly available.
Generating synthetic languages aids in the testing and validation of future computational linguistic models and methods. This thesis extends the BEAST2 phylogenetic framework to add linguistic sequence generation under multiple models. The new plugin is then used to test the effects of the phenomena of word borrowing on the inference process under two widely used phylolinguistic models.
Recent advances in corpus-based Natural Language Generation (NLG) hold the promise of being easily portable across domains, but require costly training data, consisting of meaning representations (MRs) paired with Natural Language (NL) utterances. In this work, we propose a novel framework for crowdsourcing high quality NLG training data, using automatic quality control measures and evaluating different MRs with which to elicit data. We show that pictorial MRs result in better NL data being collected than logic-based MRs: utterances elicited by pictorial MRs are judged as significantly more natural, more informative, and better phrased, with a significant increase in average quality ratings (around 0.5 points on a 6-point scale), compared to using the logical MRs. As the MR becomes more complex, the benefits of pictorial stimuli increase. The collected data will be released as part of this submission.
The task of text segmentation may be undertaken at many levels in text analysis---paragraphs, sentences, words, or even letters. Here, we focus on a relatively fine scale of segmentation, hypothesizing it to be in accord with a stochastic model of language generation, as the smallest scale where independent units of meaning are produced. Our goals in this letter include the development of methods for the segmentation of these minimal independent units, which produce feature-representations of texts that align with the independence assumption of the bag-of-terms model, commonly used for prediction and classification in computational text analysis. We also propose the measurement of texts' association (with respect to realized segmentations) to the model of language generation. We find (1) that our segmentations of phrases exhibit much better associations to the generation model than words and (2), that texts which are well fit are generally topically homogeneous. Because our generative model produces Zipf's law, our study further suggests that Zipf's law may be a consequence of homogeneity in language production.