Search results for "Language model"

A probabilistic approach to learning a visually grounded language model through human-robot interaction

2010

A Language is among the most fascinating and complex cognitive activities that develops rapidly since the early months of infants' life. The aim of the present work is to provide a humanoid robot with cognitive, perceptual and motor skills fundamental for the acquisition of a rudimentary form of language. We present a novel probabilistic model, inspired by the findings in cognitive sciences, able to associate spoken words with their perceptually grounded meanings. The main focus is set on acquiring the meaning of various perceptual categories (e. g. red, blue, circle, above, etc.), rather than specific world entities (e. g. an apple, a toy, etc.). Our probabilistic model is based on a varia…

Datorzinātne un informācijas tehnoloģijas: Datu bāzes un Informācijas sistēmas

2010

Towards Human-Bot Collaborative Software Architecting with ChatGPT

2023

Architecting software-intensive systems can be a complex process. It deals with the daunting tasks of unifying stakeholders' perspectives, designers' intellect, tool-based automation, pattern-driven reuse, and so on, to sketch a blueprint that guides software implementation and evaluation. Despite its benefits, architecture-centric software engineering (ACSE) inherits a multitude of challenges. ACSE challenges could stem from a lack of standardized processes, socio-technical limitations, and scarcity of human expertise etc. that can impede the development of existing and emergent classes of software (e.g., IoTs, blockchain, quantum systems). Software Development Bots (DevBots) trained on la…

Shrinking language models by robust approximation

2002

We study the problem of reducing the size of a language model while preserving recognition performance (accuracy and speed). A successful approach has been to represent language models by weighted finite-state automata (WFAs). Analogues of classical automata determinization and minimization algorithms then provide a general method to produce smaller but equivalent WFAs. We extend this approach by introducing the notion of approximate determinization. We provide an algorithm that, when applied to language models for the North American Business task, achieves 25-35% size reduction compared to previous techniques, with negligible effects on recognition time and accuracy.

(Re) producció, traducció i manipulació. Qüestions d'identitat i localització en un context valencià

2006

The question of identity in Valencian Catalan writing is necessarily linked bothnto the language model («model de llengua») followed by authors in original works and in translation, and to the dialogue of this model with actual local varieties. This articlenapproaches the question from the perspective of ideology in discourse, namely thenconstruction of a dialectics of belonging or not belonging, Us and Them, that is atnwork in any linguistic element and at all levels of language. In my discussion of such dialectics of cultural articulation and self-translation I present several examples of recontextualization with relevant ideological implications: intralinguistic translation between conte…

What Are the Latest Fake News in Romanian Politics? An Automated Analysis Based on BERT Language Models

2021

Social media and news outlets facilitate information sharing, while the Web is flooded by information posted online on a daily basis. However, content may be differently transmitted from case to case, based on the authors’ intentions and vocabulary, to the extent that it generates completely opposite points of view. As such, fake news have become a global phenomenon, and recent events highlight a high impact of distorted or fake information, especially on the political side, when candidates’ discourses include tendentious statements that require careful validation before completely trusting the source. This paper proposes an automated analysis of political statements in Romanian by applying…

A word prediction methodology for automatic sentence completion

2015

Word prediction generally relies on n-grams occurrence statistics, which may have huge data storage requirements and does not take into account the general meaning of the text. We propose an alternative methodology, based on Latent Semantic Analysis, to address these issues. An asymmetric Word-Word frequency matrix is employed to achieve higher scalability with large training datasets than the classic Word-Document approach. We propose a function for scoring candidate terms for the missing word in a sentence. We show how this function approximates the probability of occurrence of a given candidate word. Experimental results show that the proposed approach outperforms non neural network lang…