Showing 1–1 of 1 results for author: Blagojevic, V
-
New Methods & Metrics for LFQA tasks
Authors:
Suchismit Mahapatra,
Vladimir Blagojevic,
Pablo Bertorello,
Prasanna Kumar
Abstract:
Long-form question answering (LFQA) tasks require retrieving the documents pertinent to a query, using them to form a paragraph-length answer. Despite considerable progress in LFQA modeling, fundamental issues impede its progress: i) train/validation/test dataset overlap, ii) absence of automatic metrics and iii) generated answers not being "grounded" in retrieved documents. This work addresses ev…
▽ More
Long-form question answering (LFQA) tasks require retrieving the documents pertinent to a query, using them to form a paragraph-length answer. Despite considerable progress in LFQA modeling, fundamental issues impede its progress: i) train/validation/test dataset overlap, ii) absence of automatic metrics and iii) generated answers not being "grounded" in retrieved documents. This work addresses every one these critical bottlenecks, contributing natural language inference/generation (NLI/NLG) methods and metrics that make significant strides to their alleviation.
△ Less
Submitted 26 December, 2021;
originally announced December 2021.