Latest MethodMomentum Contrastive Learning in Speaker EmbeddingSimCLR uses samples in the current mini-batch for negative example mining. The negative sample size is constrained with the minibatch size N.8months ago (04-13)View 59commentMethodContrastive Learning in NLPContrastive learning in the latent space has recently shown great promise, which aims to make the representation8months ago (04-11)View 61commentMethodZero-Shot Stance DetectionZero-shot stance detection (ZSSD) aims to detect stance for destination unseen targets by learning stance features from known targets8months ago (04-11)View 33commentTaskStance Detection TaskStance detection aims to automatically identify one’s opinionated standpoint/attitude8months ago (04-10)View 24commentMethodCross-Lingual Domain Adaptation in NLPChen et al. (2018) proposed ADAN, an architecture based on a feed-forward neural network with three main components8months ago (04-10)View 29commentMethodDomain Adaptation in NLPSeveral works employed domain adaptation to improve performance.8months ago (04-10)View 43commentMethodMultilingual vs. Monolingual BERTDevlin et al. (2019) produced 2 BERT models, for English and Chinese. To support other languages, they trained a multilingual BERT (mBERT)8months ago (04-10)View 33commentMethodBERT Benchmarks in Large Pre-trained Language ModelsAn integral part involved in developing various PLMs is providing NLU multitask benchmarks used to demonstrate the linguistic abilities of new models and approaches8months ago (04-10)View 37commentMethodEmbedding Problems in Neural Language ModelsRecent studies on the geometric properties of contextual embedding space have observed that the distribution of embedding vectors is far from isotropic and occupies a...8months ago (04-07)View 36commentMethodText Generation of Neural Language ModelsNeural language generative models process text generation tasks as conditional language modeling8months ago (04-07)View 30comment Load More