The basic underlying assumption of machine learning (ML) models is that the training and test data are sampled from the same distribution. However, in daily practice, this assumption is often broken, i.e. the distribution of the test data changes over time, which hinders the application of conventional ML models. One domain where the distribution shift naturally occurs is text classification, since people always find new topics to discuss. To this end, we survey research articles studying open-set text classification and related tasks. We divide the methods in this area based on the constraints that define the kind of distribution shift and the corresponding problem formulation, i.e. learning with the Universum, zero-shot learning, and open-set learning. We next discuss the predominant mitigation approaches for each problem setup. We further identify several future work directions, aiming to push the boundaries beyond the state of the art. Finally, we explain how continual learning can solve many of the issues caused by the shifting class distribution. We maintain a list of relevant papers at https://github.com/Eduard6421/Open-Set-Survey.

Natural language inference (NLI), the task of recognizing the entailment relationship in sentence pairs, is an actively studied topic serving as a proxy for natural language understanding. Despite the relevance of the task in building conversational agents and improving text classification, machine translation and other NLP tasks, to the best of our knowledge, there is no publicly available NLI corpus for the Romanian language. To this end, we introduce the first Romanian NLI corpus (RoNLI) comprising 58K training sentence pairs, which are obtained via distant supervision, and 6K validation and test sentence pairs, which are manually annotated with the correct labels. We conduct experiments with multiple machine learning methods based on distant learning, ranging from shallow models based on word embeddings to transformer-based neural networks, to establish a set of competitive baselines. Furthermore, we improve on the best model by employing a new curriculum learning strategy based on data cartography. Our dataset and code to reproduce the baselines are available at https://github.com/Eduard6421/RONLI.

La prédiction de la performance des requêtes (QPP) dans le contexte de la recherche d’images basée sur le contenu reste une tâche largement inexplorée, en particulier dans le scénario de la recherche par l’exemple, où la requête est une image. Pour stimuler les recherches dans ce domaine, nous proposons la première collection de référence. Nous proposons un ensemble de quatre jeux de données (PASCAL VOC 2012, Caltech-101, ROxford5k et RParis6k) avec les performances attendues pour chaque requête à l’aide de deux modèles de recherche d’images état de l’art. Nous proposons également de nouveaux prédicteurs pré et post-recherche. Les résultats empiriques montrent que la plupart des prédicteurs ne se généralisent pas aux différents scénarios d’évaluation. Nos expériences exhaustives indiquent que l’iQPP est une référence difficile, révélant une importante lacune dans la recherche qui doit être abordée dans les travaux futurs. Nous publions notre code et nos données.