Контрфактическое увеличение данных

AbstractThis paper treats gender bias latent in word embeddings. Previous mitigation attempts rely on the operationalisation of gender bias as a projection over a linear subspace. An alternative approach is …...

AbstractSocial biases present in data are often directly reflected in the predictions of models trained on that data. We analyze gender bias in dialogue data, and examine how this bias …...

AbstractDeveloping robust NLP models that perform well on many, even small, slices of data is a significant but important challenge, with implications from fairness to general reliability. To this end, …...

AbstractRecent work has shown pre-trained language models capture social biases from the large amounts of text they are trained on. This has attracted attention to developing techniques that mitigate such …...

AbstractEntity typing aims at predicting one or more words that describe the type(s) of a specific mention in a sentence. Due to shortcuts from surface patterns to annotated entity labels …...

AbstractClass imbalance problem can cause machine learning models to produce an undesirable performance on the minority class as well as the whole dataset. Using data augmentation techniques to increase the …...

AbstractQuestion answering models commonly have access to two sources of “knowledge” during inference time: (1) parametric knowledge - the factual knowledge encoded in the model weights, and (2) contextual knowledge …...

AbstractAs the representation capability of Pre-trained Language Models (PLMs) improve, there is growing concern that they will inherit social biases from unprocessed corpora. Most previous debiasing techniques used Counterfactual Data …...