Abstract
We propose a family of metrics to assess language generation derived from population estimation methods widely used in ecology. More specifically, we use mark-recapture and maximum-likelihood methods that have been applied over the past several decades to estimate the size of closed populations in the wild. We propose three novel metrics: MEPetersen and MECAPTURE, which retrieve a single-valued assessment, and MESchnabel which returns a double-valued metric to assess the evaluation set in terms of quality and diversity, separately. In synthetic experiments, our family of methods is sensitive to drops in quality and diversity. Moreover, our methods show a higher correlation to human evaluation than existing metrics on several challenging tasks, namely unconditional language generation, machine translation, and text summarization.- Anthology ID:
- 2020.coling-main.178
- Volume:
- Proceedings of the 28th International Conference on Computational Linguistics
- Month:
- December
- Year:
- 2020
- Address:
- Barcelona, Spain (Online)
- Editors:
- Donia Scott, Nuria Bel, Chengqing Zong
- Venue:
- COLING
- SIG:
- Publisher:
- International Committee on Computational Linguistics
- Note:
- Pages:
- 1963–1977
- Language:
- URL:
- https://aclanthology.org/2020.coling-main.178
- DOI:
- 10.18653/v1/2020.coling-main.178
- Cite (ACL):
- Gonçalo Mordido and Christoph Meinel. 2020. Mark-Evaluate: Assessing Language Generation using Population Estimation Methods. In Proceedings of the 28th International Conference on Computational Linguistics, pages 1963–1977, Barcelona, Spain (Online). International Committee on Computational Linguistics.
- Cite (Informal):
- Mark-Evaluate: Assessing Language Generation using Population Estimation Methods (Mordido & Meinel, COLING 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2020.coling-main.178.pdf
- Data
- MultiNLI, SNLI