State-of-the-art NLP models can adopt shallow heuristics that limit their generalization capability (McCoy et al., 2019).
QuestEval is a reference-less metric used in text-to-text tasks, that compares the generated summaries directly to the source text, by automatically asking and answering questions.
Specifically, we propose a Multi-Branch Decoder which is able to leverage word-level labels to learn the relevant parts of each training instance.
Ranked #3 on Table-to-Text Generation on WikiBio
Evaluations on the widely used WikiBIO and WebNLG benchmarks demonstrate the effectiveness of this framework compared to state-of-the-art models.
Despite efforts to distinguish three different evaluation setups (Bekoulis et al., 2018), numerous end-to-end Relation Extraction (RE) articles present unreliable performance comparison to previous work.
This however loses most of the structure contained in the data.