Results 51 to 60 of about 248,393 (200)
Modelling Without a Modelling Language [PDF]
, 2018 Developments in computer hardware and programming languages, in this case C++, have made it feasible to write models of concurrent systems under verification in the programming language, instead of some established modelling language such as Promela. While this does not reduce the usefulness of modelling languages, it offers new possibilities that may ...
Valmari Antti, Lappalainen Vesa
openaire +2 more sources
CoRR, 2021 In almost all text generation applications, word sequences are constructed in a left-to-right (L2R) or right-to-left (R2L) manner, as natural language sentences are written either L2R or R2L. However, we find that the natural language written order is not essential for text generation. In this paper, we propose Spiral Language Modeling (SLM), a general
Yong Cao +3 more
openaire +2 more sources
Modeling adaptation with a tuple-based coordination language [PDF]
, 2012 In recent years, it has been argued that systems and applications, in order to deal with their increasing complexity, should be able to adapt their behavior according to new requirements or environment conditions.
R. Pugliese +10 more
core +1 more source
Position Models and Language Modeling [PDF]
, 2008 In statistical language modelling the classic model used is n -gram. This model is not able however to capture long term dependencies, i.e. dependencies larger than n . An alternative to this model is the probabilistic automaton. Unfortunately, it appears that preliminary experiments on the use of this model in language modelling is not yet competitive,
Arnaud Zdziobeck, Franck Thollard
openaire +1 more source
Generative Spoken Dialogue Language Modeling
Transactions of the Association for Computational Linguistics, 2023 We introduce dGSLM, the first “textless” model able to generate audio samples of naturalistic spoken dialogues. It uses recent work on unsupervised spoken unit discovery coupled with a dual-tower transformer architecture with cross-attention trained on ...
Tu Anh Nguyen +10 more
doaj +1 more source
Paraphrastic language models and combination with neural network language models [PDF]
2013 IEEE International Conference on Acoustics, Speech and Signal Processing, 2013 In natural languages multiple word sequences can represent the same underlying meaning. Only modelling the observed surface word sequence can result in poor context coverage, for example, when using n-gram language models (LM). To handle this issue, paraphrastic LMs were proposed in previous research and successfully applied to a US English ...
Liu, X, Gales, MJF, Woodland, PC
openaire +1 more source
Cognitive-Discursive Modeling of Language Nomination
Вестник Кемеровского государственного университета, 2022 Systemic and structural linguistics has no answer to the question of how one can make new words, as formulated by Prof. L. V. Shcherba. Its methodology strives to distance itself from the human factor and human activity. The present article introduces an
M. Dzh. Tagaev
doaj +1 more source
Language Models as Agent Models
Findings of the Association for Computational Linguistics: EMNLP 2022, 2022 Language models (LMs) are trained on collections of documents, written by individual human agents to achieve specific goals in an outside world. During training, LMs have access only to text of these documents, with no direct evidence of the internal states of the agents that produced them -- a fact often used to argue that LMs are incapable of ...
openaire +2 more sources
Proceedings of the National Academy of Sciences, 2017 Languages behave similarly to living species. They display diversity, differentiate in space and time, emerge and disappear. While processes of differentiation happen at a relatively slow rate with a typical timescale of the order of 1,000 years to evolve into different languages, language extinction takes place at a substantially faster rate.
Kandler, A, Steele, J
openaire +4 more sources
Frontiers in Computer ScienceRotary Positional Embedding (RoPE) is a widely used technique in Transformers, influenced by the hyperparameter theta (θ). However, the impact of varying *fixed* theta values, especially the trade-off between performance and efficiency on tasks like ...
Zhigao Huang, Musheng Chen, Shiyan Zheng
doaj +1 more source