no code implementations • 13 Dec 2023 • Wei zhang, Alexandre Salle
We present the first experiments on Native Language Identification (NLI) using LLMs such as GPT-4.
1 code implementation • 19 Aug 2019 • Alexandre Salle, Aline Villavicencio
In distributional semantics, the pointwise mutual information ($\mathit{PMI}$) weighting of the cooccurrence matrix performs far better than raw counts.
1 code implementation • 26 Apr 2019 • Alexandre Salle, Marcelo Prates
This short paper introduces an abstraction called Think Again Networks (ThinkNet) which can be applied to any state-dependent function (such as a recurrent neural network).
1 code implementation • WS 2018 • Alexandre Salle, Aline Villavicencio
The positive effect of adding subword information to word embeddings has been demonstrated for predictive models.
no code implementations • ACL 2018 • Alexandre Salle, Aline Villavicencio
Increasing the capacity of recurrent neural networks (RNN) usually involves augmenting the size of the hidden layer, with significant increase of computational cost.
1 code implementation • 3 Jun 2016 • Alexandre Salle, Marco Idiart, Aline Villavicencio
The effectiveness of both modifications is shown using word similarity and analogy tasks.
1 code implementation • ACL 2016 • Alexandre Salle, Marco Idiart, Aline Villavicencio
In this paper, we propose LexVec, a new method for generating distributed word representations that uses low-rank, weighted factorization of the Positive Point-wise Mutual Information matrix via stochastic gradient descent, employing a weighting scheme that assigns heavier penalties for errors on frequent co-occurrences while still accounting for negative co-occurrence.