The capability of the traditional semi-supervised learning (SSL) methods is far from real-
world application due to severely biased pseudo-labels caused by (1) class imbalance and …

Open-world semi-supervised learning aims at inferring both known and novel classes in
unlabeled data, by harnessing prior knowledge from a labeled set with known classes …

Open-set Semi-supervised Learning (OSSL) holds a realistic setting that unlabeled data
may come from classes unseen in the labeled set, ie, out-of-distribution (OOD) data, which …

Machine learning models deployed in the wild naturally encounter unlabeled samples from
both known and novel classes. Challenges arise in learning from both the labeled and …

Recently, pre-training methods have shown remarkable success in task-oriented dialog
(TOD) systems. However, most existing pre-trained models for TOD focus on either dialog …

Vision-language models, such as contrastive language-image pre-training (CLIP), have
demonstrated impressive results in natural image domains. However, these models often …

Pre-training methods with contrastive learning objectives have shown remarkable success
in dialog understanding tasks. However, current contrastive learning solely considers the …

We introduce MarginMatch, a new SSL approach combining consistency regularization and
pseudo-labeling, with its main novelty arising from the use of unlabeled data training …

Semi-supervised learning (SSL) methods effectively leverage unlabeled data to improve
model generalization. However, SSL models often underperform in open-set scenarios …

Abstract Semi-Supervised Learning (SSL) under class distribution mismatch aims to tackle a
challenging problem wherein unlabeled data contain lots of unknown categories unseen in …