no code implementations • WS 2019 • Elozino Egonmwan, Vittorio Castelli, Md. Arafat Sultan
We demonstrate the viability of knowledge transfer between two related tasks: machine reading comprehension (MRC) and query-based text summarization.
no code implementations • WS 2019 • Elozino Egonmwan, Yllias Chali
For better quality of generated paraphrases, we propose a framework that combines the effectiveness of two models {--} transformer and sequence-to-sequence (seq2seq).
no code implementations • WS 2019 • Elozino Egonmwan, Yllias Chali
We propose a system that improves performance on single document summarization task using the CNN/DailyMail and Newsroom datasets.