SmithE.M. x 6

bookRéférences 6

Multiple-attribute text style transfer

… Firstly, while previous approaches operate at the sentence level by assuming that every...

2019-09-22 20:07:01

SubramanianS.LampleG.SmithE.M.

Hi, my name is Martha: Using names to measure and mitigate bias in generative dialogue models

… finetuned on several dialogue datasets that were designed to impart the model with a wi...

SmithE.M.WilliamsA.

" I'm sorry to hear that": finding bias in language models with a holistic descriptor dataset

… By inspection, utterances marked as offensive tend to be those that respond to negative...

SmithE.M.KambadurM.H.M.PresaniE.

Controlling style in generated dialogue

… We do not make use of the sentiment labels from ED in this work. They might be somewhat...

SmithE.M.DinanE.GonzalezRicoD.

“I'm sorry to hear that”: Finding New Biases in Language Models with a Holistic Descriptor Dataset

… This effect is not due solely to template length, as seen when comparing longer, emotio...

SmithE.M.PresaniE.HallM.KambadurM.

Mots-clés associés