The Extraordinary Failure of Complement Coercion Crowdsourcing

Yanai Elazar, Victoria Basmov, Shauli Ravfogel, Yoav Goldberg, Reut Tsarfaty


Abstract
Crowdsourcing has eased and scaled up the collection of linguistic annotation in recent years. In this work, we follow known methodologies of collecting labeled data for the complement coercion phenomenon. These are constructions with an implied action — e.g., “I started a new book I bought last week”, where the implied action is reading. We aim to collect annotated data for this phenomenon by reducing it to either of two known tasks: Explicit Completion and Natural Language Inference. However, in both cases, crowdsourcing resulted in low agreement scores, even though we followed the same methodologies as in previous work. Why does the same process fail to yield high agreement scores? We specify our modeling schemes, highlight the differences with previous work and provide some insights about the task and possible explanations for the failure. We conclude that specific phenomena require tailored solutions, not only in specialized algorithms, but also in data collection methods.
Anthology ID:
2020.insights-1.17
Volume:
Proceedings of the First Workshop on Insights from Negative Results in NLP
Month:
November
Year:
2020
Address:
Online
Venue:
insights
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
106–116
Language:
URL:
https://aclanthology.org/2020.insights-1.17
DOI:
10.18653/v1/2020.insights-1.17
Bibkey:
Cite (ACL):
Yanai Elazar, Victoria Basmov, Shauli Ravfogel, Yoav Goldberg, and Reut Tsarfaty. 2020. The Extraordinary Failure of Complement Coercion Crowdsourcing. In Proceedings of the First Workshop on Insights from Negative Results in NLP, pages 106–116, Online. Association for Computational Linguistics.
Cite (Informal):
The Extraordinary Failure of Complement Coercion Crowdsourcing (Elazar et al., insights 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.insights-1.17.pdf
Video:
 https://slideslive.com/38940804
Data
MultiNLISNLI