Abstract
We report our efforts in identifying a set of previous human evaluations in NLP that would be suitable for a coordinated study examining what makes human evaluations in NLP more/less reproducible. We present our results and findings, which include that just 13% of papers had (i) sufficiently low barriers to reproduction, and (ii) enough obtainable information, to be considered for reproduction, and that all but one of the experiments we selected for reproduction was discovered to have flaws that made the meaningfulness of conducting a reproduction questionable. As a result, we had to change our coordinated study design from a reproduce approach to a standardise-then-reproduce-twice approach. Our overall (negative) finding that the great majority of human evaluations in NLP is not repeatable and/or not reproducible and/or too flawed to justify reproduction, paints a dire picture, but presents an opportunity for a rethink about how to design and report human evaluations in NLP.
Original language | English |
---|---|
Title of host publication | The Fourth Workshop on Insights from Negative Results in NLP |
Editors | Shabnam Tafreshi, Arjun Akula, João Sedoc, Aleksandr Drozd, Anna Rogers, Anna Rumshisky |
Place of Publication | Dubrovnik, Croatia |
Publisher | Association for Computational Linguistics |
Pages | 1-10 |
Number of pages | 10 |
DOIs | |
Publication status | Published - 1 May 2023 |
Event | Insights 2023 : The Forth Workshop on Insights from Negative Results in NLP - Dubrovnik, Croatia Duration: 2 Jun 2023 → 6 Jun 2023 |
Workshop
Workshop | Insights 2023 : The Forth Workshop on Insights from Negative Results in NLP |
---|---|
Country/Territory | Croatia |
City | Dubrovnik |
Period | 2/06/23 → 6/06/23 |