Reproduce “An End-to-End Network for Emotion-Cause Pair Extraction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Aaditya Singh, Shreeshail Hingane, Saim Wani, Ashutosh Modi
The authors introduce end-to-end emotion-cause pair extraction (ECPE), predicting emotion clauses, cause clauses, and their links together. A hierarchical bidirectional LSTM architecture lets emotion predictions inform cause encoding and trains pair and clause classifiers jointly. On an adapted English novel corpus, it achieves pair-level F1 of 50.17 versus 43.67 for a two-stage baseline, a result limited to that benchmark.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.