Showing cs.CLShow all
2 papers · 1 filter
cs.CL2023
SQUARE: Automatic Question Answering Evaluation using Multiple Positive and Negative References
Matteo Gabburo, Siddhant Garg, Rik Koncel Kedziorski +1
Evaluation of QA systems is very challenging and expensive, with the most reliable approach being human annotations of correctness of answers for questions. Recent works (AVA, BEM)…
cs.CL2023
Learning Answer Generation using Supervision from Automatic Question Answering Evaluators
Matteo Gabburo, Siddhant Garg, Rik Koncel-Kedziorski +1
Recent studies show that sentence-level extractive QA, i.e., based on Answer Sentence Selection (AS2), is outperformed by Generation-based QA (GenQA) models, which generate answers…