A two-module denoising model for multimodal aspect-based sentiment analysis reports slight F1 gains on Twitter-15/17, with the most relevant baseline missing from the experiments.
Dynamically Composing Domain-Data Selection with Clean-Data Selection by "Co-Curricular Learning" for Neural Machine Translation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Noise and domain are important aspects of data quality for neural machine translation. Existing research focus separately on domain-data selection, clean-data selection, or their static combination, leaving the dynamic interaction across them not explicitly examined. This paper introduces a "co-curricular learning" method to compose dynamic domain-data selection with dynamic clean-data selection, for transfer learning across both capabilities. We apply an EM-style optimization procedure to further refine the "co-curriculum". Experiment results and analysis with two domains demonstrate the effectiveness of the method and the properties of data scheduled by the co-curriculum.
fields
cs.CV 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
A Dual-Module Denoising Approach with Curriculum Learning for Enhancing Multimodal Aspect-Based Sentiment Analysis
A two-module denoising model for multimodal aspect-based sentiment analysis reports slight F1 gains on Twitter-15/17, with the most relevant baseline missing from the experiments.