Модель-Агностикалық Мета-Оқу
Related Papers
Few-Shot Representation Learning for Out-Of-Vocabulary Words
AbstractExisting approaches for learning word embedding often assume there are sufficient occurrences for each word in the corpus, such that the representation of words can be accurately estimated from their …...
Personalizing Dialogue Agents via Meta-Learning
AbstractExisting personalized dialogue models use human designed persona descriptions to improve dialogue consistency. Collecting such descriptions from existing dialogues is expensive and requires hand-crafted feature designs. In this paper, we …...
Meta-Reinforced Multi-Domain State Generator for Dialogue Systems
AbstractA Dialogue State Tracker (DST) is a core component of a modular task-oriented dialogue system. Tremendous progress has been made in recent years. However, the major challenges remain. The state-of-the-art …...
Boosting Natural Language Generation from Instructions with Meta-Learning
AbstractRecent work has shown that language models (LMs) trained with multi-task instructional learning (MTIL) can solve diverse NLP tasks in zero- and few-shot settings with improved performance compared to prompt …...
Decomposed Meta-Learning for Few-Shot Named Entity Recognition
AbstractFew-shot named entity recognition (NER) systems aim at recognizing novel-class named entities based on only a few labeled examples. In this paper, we present a decomposed meta-learning approach which addresses …...
Meta-Learning for Fast Cross-Lingual Adaptation in Dependency Parsing
AbstractMeta-learning, or learning to learn, is a technique that can help to overcome resource scarcity in cross-lingual NLP problems, by enabling fast adaptation to new tasks. We apply model-agnostic meta-learning …...
MetaPrompting: Learning to Learn Better Prompts
AbstractPrompting method is regarded as one of the crucial progress for few-shot nature language processing. Recent research on prompting moves from discrete tokens based “hard prompts” to continuous “soft prompts”, …...