Search

Generative Artificial Intelligence has grown exponentially as a result of Large Language Models (LLMs). This has been possible because of the impressive performance of deep learning methods created within the field of Natural Language Processing (NLP) and its subfield Natural Language Generation (NLG), which is the focus of this paper. Within the growing LLM family are the popular GPT-4, Bard and more specifically, tools such as ChatGPT have become a benchmark for other LLMs when solving most of the tasks involved in NLG research. This scenario poses new questions about the next steps for NLG and how the field can adapt and evolve to deal with new challenges in the era of LLMs. To address this, the present paper conducts a review of a representative sample of surveys recently published in NLG. By doing so, we aim to provide the scientific community with a research roadmap to identify which NLG aspects are still not suitably addressed by LLMs, as well as suggest future lines of research that should be addressed going forward.

The “Conscious Natural Text Generation” project (CORTEX, grant ref. PID2021-123956OB-I00) investigates how Natural Language Generation (NLG) architectures can be improved so that they can exploit external and commonsense knowledge, and integrate it in a controllable manner (i.e., determining what knowledge to include and how to include it) when generating new text. Our main scientific objective is, therefore, to investigate and propose novel NLG architectures that integrate different types of knowledge to automatically produce reliable, truthful and quality texts, whose information complies with the principles of objectivity and plurality, thereby minimizing bias or manipulated content. The new generation of NLG systems obtained from CORTEX will significantly improve the semantic quality of the generated text, preventing, among other phenomena, the inclusion of invented facts that do not match reality (i.e., hallucination).

We present a data-driven approach to discover and extract patterns in textual genres with the aim of identifying whether there is an interesting variation of linguistic features among different narrative genres depending on their respective communicative purposes. We want to achieve this goal by performing a multilevel discourse analysis according to (1) the type of feature studied (shallow, syntactic, semantic, and discourse-related); (2) the texts at a document level; and (3) the textual genres of news, reviews, and children’s tales. To accomplish this, several corpora from the three textual genres were gathered from different sources to ensure a heterogeneous representation, paying attention to the presence and frequency of a series of features extracted with computational tools. This deep analysis aims at obtaining more detailed knowledge of the different linguistic phenomena that directly shape each of the genres included in the study, therefore showing the particularities that make them be considered as individual genres but also comprise them inside the narrative typology. The findings suggest that this type of multilevel linguistic analysis could be of great help for areas of research within natural language processing such as computational narratology, as they allow a better understanding of the fundamental features that define each genre and its communicative purpose. Likewise, this approach could also boost the creation of more consistent automatic story generation tools in areas of language generation.

Communicative intentions are one of the linguistic elements that usually determine the content of any message we want to express in our social interactions. With the purpose of contributing to the improvement of natural language processing systems, this thesis aims to create a communicative intention annotation scheme based on the taxonomy presented in the Speech Act Theory. In this way, language processing tools could consider communicative intentions as a starting point to help classify any message and its content depending first on the intention it reflects. To do so, the scheme will be created with the help of an already annotated corpus of Spanish tweets and subsequently evaluated by external annotators so that we can confirm the appropriateness and reliability of the tagged intentions before applying the scheme to an NLP system. Thus, it will be possible to check up to which point communicative intentions can improve the identification of the purpose of a message in an already created NLP system so that we can gain more linguistic information from any text automatically.

Communicative intentions are one of the linguistic elements that usually determine the content of any text or message we want to express in our communicative interactions. With the purpose of contributing to the improvement of natural language generation systems, so that they can take the communicative intention as one of the starting points that will determine the structure and content of the message generated, the aim of this project is to create a communicative intentions annotation scheme based on the taxonomy presented in the Speech Act Theory. To do so, the scheme will be created with the help of a linguistic corpus and subsequently tested within a natural language generation system. In this way, it will be possible to check up to which point communicative intentions improve the planning stage of the text to be generated automatically, guiding the rest of decisions to be made by the system in order to create automatic messages with more similar results to any manually created text.

En este informe científico se identifican los principales retos científicos en los cuales está enmarcado el grupo de investigación en procesamiento del lenguaje y sistemas de información (GPLSI) de la Universidad de Alicante (UA). El objetivo de este informe es establecer un plan de investigación basado en la experiencia y participación en proyectos de investigación, y subrayar los principales retos de futuro que nos permitan definir nuevos proyectos de investigación y desarrollo en los próximos años. La Generación Automática de Lenguaje Natural (GLN) es una tarea compleja que se apoya en numerosos recursos y herramientas del Procesamiento del Lenguaje Natural (PLN). La GLN comienza a dar resultados satisfactorios, al menos para ciertos dominios y objetivos, como se puede ver de los resultados obtenidos por parte del GPLSI. No obstante, requiere un importante esfuerzo investigador. En este artículo planteamos, desde la perspectiva que da la investigación realizada hasta ahora en grupo de investigación Procesamiento Natural y Sistemas de Información (GPLSI), futuras líneas de investigación en GLN, incorporando diferentes disciplinas y metas.

The analysis of discourse and the study of what characterizes it in terms of communicative objectives is essential to most tasks of Natural Language Processing. Consequently, research on textual genres as expressions of such objectives presents an opportunity to enhance both automatic techniques and resources. To conduct an investigation of this kind, it is necessary to have a good understanding of what defines and distinguishes each textual genre. This research presents a data-driven approach to discover and analyze patterns in several textual genres with the aim of identifying and quantifying the differences between them, considering how language is employed and meaning expressed in each particular case. To identify and analyze patterns within genres, a set of linguistic features is first defined, extracted and computed by using several Natural Language Processing tools. Specifically, the analysis is performed over a corpora of documents—containing news, tales and reviews—gathered from different sources to ensure an heterogeneous representation. Once the feature dataset has been generated, machine learning techniques are used to ascertain how and to what extent each of the features should be present in a document depending on its genre. The results show that the set of features defined is relevant for characterizing the different genres. Furthermore, the findings allow us to perform a qualitative analysis of such features, so that their usefulness and suitability is corroborated. The results of the research can benefit natural language discourse processing tasks, which are useful both for understanding and generating language.

El campo de la subtitulación siempre ha sido concebido como un mero elemento externo a la obra cinematográfica que debe pasar desapercibido lo máximo posible para poder cumplir su doble función: interlingüística al ser un tipo de mediación entre diferentes idiomas e intralingüística cuando hablamos de subtítulos enfocados a la accesibilidad de un producto audiovisual; todo ello sin afectar al completo visionado de la película. Sin embargo, recientemente se han dado varios casos en los que dichos subtítulos pasan a tener un papel principal y primordial para poder comprender la historia que nos presenta cada película. Es por ello que este estudio pretende analizar dicha innovación en la estructura de los subtítulos e investigar las posibles características de esta rama tan reciente que abre el abanico de posibilidades de la traducción de los subtítulos para poder llegar de manera más directa al público, además de por supuesto transmitir el mensaje de la obra. Para ello se tomará como corpus de análisis las tres primeras temporadas de la exitosa serie británica de la BBC Sherlock, que ha servido de referente para estudiar las diferentes posibilidades de la llamada subtitulación creativa.