How Many Layers and Why? An Analysis of the Model Depth in Transformers

no code implementations • ACL 2021 • Antoine Simoulin, Benoit Crabb{\'e}

In this study, we investigate the role of the multiple layers in deep transformer models.

Specificity

FrSemCor: Annotating a French Corpus with Supersenses

no code implementations • LREC 2020 • Lucie Barque, Pauline Haas, Richard Huyghe, Delphine Tribout, C, Marie ito, Benoit Crabb{\'e}, Vincent Segonne

French, as many languages, lacks semantically annotated corpus data.

Variable beam search for generative neural parsing and its relevance for the analysis of neuro-imaging signal

no code implementations • IJCNLP 2019 • Benoit Crabb{\'e}, Murielle Fabre, Christophe Pallier

This paper describes a method of variable beam size inference for Recurrent Neural Network Grammar (rnng) by drawing inspiration from sequential Monte-Carlo methods such as particle filtering.

Multilingual discriminative lexicalized phrase structure parsing

no code implementations • EMNLP 2015 • Benoit Crabb{\'e}

Dependency Parsing Feature Engineering

Dependency length minimisation effects in short spans: a large-scale analysis of adjective placement in complex noun phrases

no code implementations • IJCNLP 2015 • Kristina Gulordava, Paola Merlo, Benoit Crabb{\'e}

An LR-inspired generalized lexicalized phrase structure parser

no code implementations • COLING 2014 • Benoit Crabb{\'e}

Dependency Parsing

Towards a treebank of spoken French (Vers un treebank du fran\ccais parl\'e) [in French]

no code implementations • JEPTALNRECITAL 2013 • Anne Abeill{\'e}, Benoit Crabb{\'e}

Ubiquitous Usage of a Broad Coverage French Corpus: Processing the Est Republicain corpus

no code implementations • LREC 2012 • Djam{\'e} Seddah, C, Marie ito, Benoit Crabb{\'e}, Enrique Henestroza Anguiano

In this paper, we introduce a set of resources that we have derived from the EST R{\'E}PUBLICAIN CORPUS, a large, freely-available collection of regional newspaper articles in French, totaling 150 million words.

Dependency Parsing Lemmatization +1