Вспомогательные задачи
Related Papers
When is multitask learning effective? Semantic sequence prediction under varying data conditions
AbstractMultitask learning has been applied successfully to a range of tasks, mostly morphosyntactic. However, little is known on when MTL works and whether there are data characteristics that help to …...
Syntax-aware Semantic Role Labeling without Parsing
AbstractIn this paper we focus on learning dependency aware representations for semantic role labeling without recourse to an external parser. The backbone of our model is an LSTM-based semantic role …...
W-RST: Towards a Weighted RST-style Discourse Framework
AbstractAiming for a better integration of data-driven and linguistically-inspired approaches, we explore whether RST Nuclearity, assigning a binary assessment of importance between text segments, can be replaced by automatically generated, …...
MT-Speech at SemEval-2022 Task 10: Incorporating Data Augmentation and Auxiliary Task with Cross-Lingual Pretrained Language Model for Structured Sentiment Analysis
AbstractSentiment analysis is a fundamental task, and structure sentiment analysis (SSA) is an important component of sentiment analysis. However, traditional SSA is suffering from some important issues: (1) lack of …...