International audienceThis paper describes a study on gender bias in French/English neural machine translation (MT) systems. We introduce a controlled corpus to measure the intensity of such biases in the two translation directions (from and into English). This corpus also allows us to investigate the information flow in a encoder-decoder architecture and to identify how gender information can be transfered between languages. Considering both probing as well as interventions on the internal representations of the MT system, we show that gender information is encoded in all token representations built by the encoder and the decoder and that there are multiple paths to transfer gender. MOTS-CLÉS : biais de genre, traduction automatique neuron...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
Multilingual Neural Machine Translation architectures mainly differ in the amount of sharing modules...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
Multilingual Neural Machine Translation architectures mainly differ in the amount of sharing modules...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper describes a study on gender bias in French/English neural machine t...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper is a blueprint of a current study in the making on gender bias in F...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
International audienceThis paper aims at identifying the information flow in state-of-the-art machin...
Multilingual Neural Machine Translation architectures mainly differ in the amount of sharing modules...