New Protocols and Negative Results for Textual Entailment Data Collection

Samuel R. Bowman, Jennimaria Palomaki, Livio Baldini Soares, Emily Pitler

Semantics: Sentence-level Semantics, Textual Inference and Other areas Long Paper

Gather-5B: Nov 18, Gather-5B: Nov 18 (18:00-20:00 UTC) [Join Gather Meeting]

You can open the pre-recorded video in a separate window.

Abstract: Natural language inference (NLI) data has proven useful in benchmarking and, especially, as pretraining data for tasks requiring language understanding. However, the crowdsourcing protocol that was used to collect this data has known issues and was not explicitly optimized for either of these purposes, so it is likely far from ideal. We propose four alternative protocols, each aimed at improving either the ease with which annotators can produce sound training examples or the quality and diversity of those examples. Using these alternatives and a fifth baseline protocol, we collect and compare five new 8.5k-example training sets. In evaluations focused on transfer learning applications, our results are solidly negative, with models trained on our baseline dataset yielding good transfer performance to downstream tasks, but none of our four new methods (nor the recent ANLI) showing any improvements over that baseline. In a small silver lining, we observe that all four new protocols, especially those where annotators edit *pre-filled* text boxes, reduce previously observed issues with annotation artifacts.
NOTE: Video may display a random order of authors. Correct author list is at the top of this page.

Connected Papers in EMNLP2020

Similar Papers