Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2022 · 18 citations · 26 references
ComputingComputer ScienceComputational LinguisticsMachine LearningNatural Language ProcessingActive LearningAnnotationCoreference ResolutionCorpus AnnotationLanguage ModelingNamed Entity RecognitionActive Learning For ClassificationActive Learning With Deep Neural NetworksModel UncertaintyNeural Coreference ResolutionSampling StrategiesSampling UncertaintySpan-based Coreference ModelsUncertainty Sampling
Neural coreference resolution models trained on one dataset may not transfer to new, lowresource domains. Active learning mitigates this problem by sampling a small subset of data for annotators to label. While active learning is well-defined for classification tasks, its application to coreference resolution is neither well-defined nor fully understood. This paper explores how to actively label coreference, examining sources of model uncertainty and document reading costs. We compare uncertainty sampling strategies and their advantages through thorough error analysis. In both synthetic and human experiments, labeling spans within the same document is more effective than annotating spans across documents. The findings contribute to a more realistic development of coreference resolution models.
26
Active Learning Literature Survey
Burr Settles · Minds at UW (University of Wisconsin) · 2009 · 4.8K citations · Full text
A model-theoretic coreference scoring scheme
Marc Vilain, John D. Burger, John Aberdeen et al. · 1995 · 682 citations · Full text
On coreference resolution performance metrics
Xiaoqiang Luo · 2005 · 540 citations · Full text
A sequential algorithm for training text classifiers
David Lewis · ACM SIGIR Forum · 1995 · 399 citations